Holding It Together? Professional Perspectives on the Role of Relationships when Relocating Young People due to Extra-familial Harm

When young people come to harm in extra-familial contexts, professionals may move them a distance from their home community to protect them, and in doing so disrupt relationships in which they have encountered harm. However, relocations can also fracture young people’s protective relationships with family, peers, and professionals; relationships that have been positioned as targets for intervention in cases of extra-familial harm. The extent to which these relationships are considered during relocations is under-explored. Utilising semi-structured interviews with 16 social work professionals in England and Wales, we assessed their accounts of using relationships prior to, during, and following relocations in cases of extra-familial harm. Three themes emerged: using relationships during relocations to provide consistency, to collaborate, and to create safety. Professional accounts prioritised young people’s relationships with practitioners, over relationships with families, peers, and their wider communities, when using/seeking opportunities to offer consistency and to collaborate on safety plans. They also depicted a struggle to engage with the complex web of family, peer, and community relationships associated to young people’s protection in both their home communities and those they had been moved to; relationships that were critical for creating safety. Implications for practice and future research are discussed, highlighting the potential merits of offering integrated research and practice frameworks that hold together young people’s relationships with families, peers, communities, and professionals, in response to extra-familial harm.


Introduction
During adolescence, young people's exposure to harm in extra-familial, as opposed to familial, contexts increases (Brandon et al., 2020). In England, growing concerns about this issue resulted in the UK Government grouping the physical, sexual, and emotional abuse that occurs in extra-familial contexts under a definition of 'extra-familial threats'; stating in national guidelines that these forms of harm may warrant a child protection response: As well as threats to the welfare of children from within their families, children may be vulnerable to abuse or exploitation from outside their families. These extra-familial threats might arise at school and other educational establishments, from within peer groups, or more widely from within the wider community and/ or online. These threats can take a variety of different forms … including: exploitation by criminal gangs and organised crime groups such as county lines; trafficking, online abuse; sexual exploitation and the influences of extremism leading to radicalisation (HM Government, 2018: 22) Whilst there is much that may be qualitatively different about these forms of harm, what they have in common is that they all often occur through relationships or interactions with adults and peers unconnected to young people's parents/carers.
International research into different forms of extra-familial harm (EFH) suggests that relationships form a key part of protective responses. In particular: the role of 'trusted relationships' between young people and professionals (Lefevre et al., 2017); safety provided through young people's peer relationships (Fagan & Catalano, 2012); and the ongoing role of parent-child relationships in offering protection during adolescence (Pike & Langham, 2019).
However, the contextual and social dynamics of EFH mean that in many countries around the world, professionals will relocate young people away from their home communities (and any protective relationships that they may have within them), to extricate them from extra-familial contexts and relationships where they have come to harm (Aussemsa et al., 2020;Dierkhisinga et al., 2020;Ellis, 2018;Sapiro et al., 2016;Shuker, 2013).
In this paper, we analyse data from interviews with professionals from England and Wales to explore how they describe using and protecting relationships when relocating young people affected by EFH. Through this analysis, we ask to what extent their discussions reflect the aforementioned evidence on the role of relationships (between young people and professionals, their peers, and their parents) in creating safety for young people affected by EFH. We discuss our findings and recommend the development of a framework for safeguarding young people that is capable of holding all relationships in mind when responding to EFH.

Extra-familial Harm
A range of 'harm-types' have been grouped under a definition of 'extra-familial threats' within England's child protection guidelines (as noted in the 'Introduction') -referred to as EFH in this paper. EFH includes instances where young people: • Are groomed to believe they are in a romantic relationship with adults who are exploiting them (Eaton & Holmes, 2017;Pearce, 2009) • Are in coerced or violently 'enforced' relationships with the people who exploit them, sexually or criminally (Firmin, 2017;Turner et al., 2019) • Experience abuse in their first romantic or 'dating' relationships with similarly aged peers (Barter, 2009;Foshee et al., 2014) • Experience acts of weapon-enabled violence in their neighbourhoods (Firmin, 2017;Bulanda and McCrea, 2013) • Are sexually harassed, and bullied in other ways, within educational and sports settings (Coker, 2017;Lloyd, 2018;Miller et al., 2013) By their interpersonal nature, all these forms of EFH are relational; occurring in social interactions between young people and their peers or adults unconnected to their families (Barter, 2009;Cockbain, 2018;Hallett et al., 2019;Hill, 2019;Pearce, 2009).

Relationships as Part of the Professional Response to Extra-familial Harm
A range of research studies and associated practice traditions have been drawn upon to develop responses to EFH. Developments in England have focussed on three features of EFH that national social work systems have struggled to accommodate (Hanson and Holmes, 2014). Firstly, that EFH occurs in largely extra-familial, as opposed to in parent-child, relationships; England's contemporary child protection system was designed to engage with the latter not the former set of relationships. Secondly, that EFH tends to impact young people during adolescence, as opposed to earlier childhood; those working in England's child protection system have struggled to form relationships with young people themselves (at a time when their desire for autonomy increases) and not just their parents/carers (Firmin, 2020;Child Safeguarding Practice Revew Panel, 2020;Pearce, 2013). Thirdly, that EFH can undermine parent-child relationships that would have been otherwise relatively protective. For the purposes of this paper, we present three areas of research and associated practices that have been used in England to design responses to EFH (and to the aforementioned challenges), whilst also centring relationships as a part of a protective response.

Responding to EFH Through Trusted Relationships Between Professionals and Young People
Young people routinely state that having access to an adult that they trust is critical to protecting them from EFH (Cossar et al., 2013;Ellis, 2018;Hudek, 2018;Lefevre et al., 2017). When exploring this idea, researchers have concluded that the ability of practitioners to build 'trusted relationships' with young people affected by EFH serves two protective functions.
Firstly, many responses to EFH require the voluntary engagement of young people in support services (Brandon et al., 2020;Hanson and Holmes, 2014), as opposed to via a statutory child protection plan (with which their parent/carers may be required to comply). This requirement for voluntary engagement coincides with a time of significant change in an individual's development -adolescence (Hanson and Holmes, 2014). One feature of this developmental period is an increasing desire for autonomy; to have some choice and control over one's life (Coleman, 2011). Collaboration with young people, therefore, is likely to be a key feature of service responses to EFH and has been long situated as a key principle in the delivery of youth work services (National Youth Agency, 2020). However, experiences of EFH can create barriers to collaboration, and therefore voluntary engagement. EFH can result in trauma, incite fear in young people about others who pose a risk to their safety, or mean a young person behaves in ways that pose a risk to others as well as themselves (Firmin, Wroe and Lloyd, 2019). By building trusted relationships with young people, practitioners have been found to overcome these barriers (Lefevre et al., 2017;McGuire, 2018;Scott and Botcherby, 2017), and on occasions have created an 'alliance' (Sturrock, 2012) with a young person to co-develop shared safety plans.
The second identified function of using trusted relationships in response to EFH has been to bridge access to interventions and support more widely (Lefevre et al., 2017;Lewing et al., 2018;McGuire, 2018;National Youth Agency, 2020;Scott, Botcherby and Ludvigsen, 2017). Due to the stability provided by a trusted relationship with one professional, young people may feel able to re-engage with education or training opportunities provided by other services; access housing or employment support; formally report experiences of EFH; or attend health appointments for example (Brennan & McElvaney, 2020;Cossar et al., 2013).
As such, researchers have recommended that practitioners in England build trusted relationships with young people affected by EFH (Lefevre et al., 2017;Kohli et al., 2015, National Youth Agency, 2020, and that the wider systems in which these practitioners work create the conditions to make this feasible (Hanson and Holmes, 2013;Scott, Botcherby and Ludvigsen, 2017).

Responding to EFH via Safer Peer and Community Relationships
EFH often occurs in community or school contexts and/or through peer relationships. Research in the USA, Australia, and the UK, amongst others, have highlighted interventions that build safety in those contexts as part of a response to EFH (Anon, 2020;Fagan & Catalano, 2012).

3
These interventions are varied. Some utilise situational crime prevention theories to increase guardianship in areas associated with violence (Smallbone et al., 2012). When taking this approach, practitioners identify: 'place-managers' who can influence the nature of a space; and 'guardians' and 'restrainers' who can respectively look out for the welfare of young people and disrupt the behaviours of those who want to harm them in that space. An alternative situational/contextual approach has been the use of 'bystander intervention' principles in schools and colleges to create protective peer cultures (Coker, 2017;Foshee et al., 2014;); or peer support networks in community settings (Bulanda & McCrea, 2013;Murray et al., 2016). These interventions capitalise on peer-influence during adolescence and seek to create pro-social norms in the spaces where young people spend their time.
For these approaches, practitioners utilise existing relationships or create new ones in contexts associated with EFH. These relationships are used alongside a suite of wider activities (such as changing the physical design of places) to target the social conditions in which EFH is more likely. As such, they differ from approaches to build trusted practitioner-young person relationships; they target the contexts in which EFH occurs rather than the individual affected by it. They are still relational but engage a broader network of relationships in creating safety. Peer interventions work with the dynamics of adolescent relationships, utilising, rather than disrupting, peer bonds that characterise this stage of development. Wider interventions with extra-familial contexts seek to create the social conditions in which peer and community relationships can flourish -potentially safeguarding far more young people than the small number who come to the attention of statutory services due to EFH.
Research into the benefits of contextual interventions has been drawn upon to design and test contextual approaches to safeguarding young people from EFH England (Anon, 2020). These approaches have featured social work assessment and intervention with peer groups, schools, and locations where EFH occurs -and not just with the individual young people affected. Child protection policy in England, Wales, and Scotland has been revised to promote this direction of travel, with all recommending that social work responses to EFH target environmental and contextual drivers (HM Government, 2018;Scottish Government, 2020;Welsh Government, 2019).

Responding to EFH via Relational Work with Parents and Carers
The focus of child and family social work in the UK is orientated around familial relationships, in the context of assessing and supporting parental capacity. A 'relational turn' in social work (Ruch, 2010) -in response to New Public Management's proceduralised requirements (Trevithick, 2014) -has re-invigorated the relationship between practitioners and families around a joint emphasis on social/ practical and emotional/psychological needs (Ruch, 2010). In relationship-based practice (RBP), families are seen as relational systems (Boddy et al., 2011) with the needs of young people and parent/carers interrelated (Munro, 2011). Practice then focusses on building a containing and therapeutic (Howe, 2010) practitioner/ carer-parent relationship, which, it is hoped, has an isomorphic parallel effect on the parent-young person relationship (Walsh, 2010).

3
Focussing on parent/carer-child relationship as a means of bringing about safety in extra-familial contexts, however, is likely to be challenged by the influence of peers, in adolescence, over familial relationships (Coleman, 2011). Threats of serious harm for leaving a group, the impact of grooming, or simply a disconnect between life outside and within the home can all undermine the parent/carerchild relationship (Firmin, 2017(Firmin, , 2020Hudeck, 2018;Pike & Langham, 2019). Parent/carers of young people who have experienced EFH have described 'losing' their child to street, or peer, influences, and feeling unable to keep their child safe (Pike & Langham, 2019). Therefore, punitive attempts to focus on how a parent can safeguard their child from harm at school or in their communities are likely to be ineffective (Firmin, 2017(Firmin, , 2020Child Safeguarding Practice Revew Panel, 2020;UK Anti-Slavery Commissioner, 2020) and perpetuate a cycle of mistrust between professionals and parents. Consequently, calls have been made to re-think the relationship between parents and practitioners in cases of EFH, such that parents can be partners alongside practitioners, capitalising on their shared motivation to bring about safety (Pike & Langham, 2019;Scott & Botcherby, 2017).
The concept of 'push and pull factors' (Biehal and Wade, 2000) -which has risen in prominence within EFH safeguarding -is a framework which acknowledges that harm can originate within or outside the home and can interrelate. It invites analysis of how challenges at home -such as domestic abuse or parental substance misuse (Hallett, 2017; Mason-Jones & Loggie, 2020) -can 'push' a young person to spend time outside their home, thus increasing their exposure to extra-familial harm. Equally, young people might be 'pulled' into unsafe extrafamilial contexts through coercive relationships outside their family settings. Whilst this can happen with no pre-existing familial issues (Firmin, 2020;Pike & Langham, 2019), 'pull' factors can however cause difficulties at home, as tensions and worries increase. Assessing and responding to the cause and effects of EFH harm in such situations requires relational sensitivity from practitioners.
Consequently, working with families affected by EFH using RBP requires appreciation of complex relational system. Oftentimes parent/carers and young people are at odds with one another (Firmin, 2017(Firmin, , 2020Hudeck, 2018;Pike & Langham, 2019), within the context of a historic relationships with services which families have experienced as punitive (Shuker and Ackerley, 2017). Relationship-based support may therefore need to address this legacy, alongside defusing tension between parents and young people and increasing parent/carers' understanding of EFH and exploitation.

Relationships and Social Care Responses to EFH -Three Relational Tracks
The three areas of research, and associated practice approaches, summarised above differ from each other in two ways. As depicted in Fig. 1, they each prioritise different relationships in response to EFH, and they each address different challenges of England's child protection response to EFH. One prioritises building trustedpractitioner-young person relationships to address EFH, and so seeks to improve how child protection systems, and social workers within them, voluntarily engage with adolescents. One utilises/creates relationships in the peer groups, schools, and 1 3 publics places where EFH occurs, and so improves how child protection systems, and social workers within them, respond to contexts beyond family homes. And one is focussed on protecting/rebuilding relationships between young people and their parent/carers to diminish the 'push' factors that may increase a young person's exposure to EFH; this reframes how child protection systems position and approach parents (from the target of their interventions to partners in creating safety).
All three areas of research and practice appear relevant to EFH; and, when considered collectively, they provide a sound evidence-base to suggest that relationships are a critical component of any service response to EFH.

Extra-familial Harm and Relocations
The relational features of many responses to EFH (detailed above) are potentially compromised when young people are relocated for their own safety (Shuker, 2013;Anon, 2019). When EFH occurs in neighbourhood and school settings -or within peer or wider community networks -research from the UK and other parts of Europe, Australia, and the USA has evidenced the use of out-of-home residential or foster care, secure placements, and other forms of relocation by statutory agencies (Aussemsa et al., 2020;Anon, 2019;Dierkhisinga et al., 2020;Ellis, 2018;Sapiro, 2016;Shuker, 2013). This approach removes young people from those who pose a risk to their safety and is intended to prevent them from going 'missing' and running to relationships/networks that are considered 'harmful'.
Relocations nearly always disrupt or dismantle relationships (Shuker, 2013). In some cases, this has been offered as a reason to relocate: it is an intervention that breaks relationships in which young people experience harm (Anon, 2019; Sapiro et al., 2016). Yet researchers have also noted that relocations can have a sledgehammer effect; disrupting other protective relationships that young people have with family members, peers, and/or professionals, as well as any relationships in which they encounter harm (Scott & Botcherby, 2017;Shuker, 2013). Policymakers, politicians, and researches in England have also noted that when it comes to criminal networks, relocations do not disrupt relationships sufficiently and young people can be found and harmed wherever they are placed (Home Affairs Select Committee, 2019). Concerns about the risks that relocations pose to young people's protective relationships (and queries about the extent to which they provide sustainable protection) have led some organisations to trial service innovations that end relocations or significantly reduce their use (Lushey, et al., 2017;Scott & Botcherby, 2017). Other research has explored methods of recreating relational safety once a young person has been relocated, for example, a trial of 'specialist' foster placements for young people who had been sexually exploited aimed to create a trusted relationship between a young person and their foster carer (Shuker, 2013). The foster-careryoung person relationship is then positioned as a route to creating 'relational-safety' -a secure base for a young person in a time of crisis (Shuker, 2013).

This Study
Whether relocations are used to disrupt relationships associated with EFH, or avoided to protect relationships affected by EFH, relational service responses to EFH inform the relocation debate and vice versa. In this study, we examine social work accounts of using relocations in response to EFH through a lens of the three relational approaches to EFH intervention introduced previously. In particular we ask: -What relationships (practitioner-young person, peer/community-young person, or parent/family-young person) do social work professionals prioritise when describing their use of relocation in response to EFH? -How do social workers describe using relationships prior to, during, and after relocations in cases of EFH? -In what ways do the answers to the two questions above reflect research, practice, and policy developments that have promoted interventions with EFH via practitioner-young person, peer/community-young person, or parent/family-young person relationships?
In doing so, we explore how to align relational and relocation-based responses to EFH and offer the foundations for developing an integrated relational intervention for EFH in the future.

Methodology
This study is part of a 3-year (2019-2021) mixed-methods project, titled Securing Safety, which explored the rate, cost, and impact of relocation (out-of-area and secure placements) as an intervention in cases of EFH. The first phase of the Securing Safety 1 3 project engaged 20 children's social care departments (referred to as sites hereafter) in England and Wales via surveys and follow-up interviews to understand the rate at which they used relocations in response to EFH and their rationale for doing so. During the second phase of Securing Safety, researchers interviewed young people, parents, and practitioners who had been involved in relocations to build case studies of the conditions in which relocations were (or were not) an effective intervention in cases of EFH. This study reports on analysis from data collected during the first phase of the Securing Safety project.

Dataset
Twenty participating sites were invited to submit survey data on the rate at which they used relocations in response to EFH in the month of September 2019 -13 were able to do so. Professionals (n = 16) from each of the 13 sites who coordinated the survey return and an additional one site who was unable to submit survey data, participated in a total of 15 follow-up interviews via video call with the project's lead researcher. During these interviews, participants provided a context to their survey submissions (or lack thereof), detailing the various rationales of their service for using relocations (or not) in response to EFH (Table 1).
A semi-structured interview schedule was used covering three topics of discussion: 1. How the site established and reported the data on relocations and EFH that they submitted to the survey. 2. The professionals' view on why and in what circumstances a relocation would be considered in their site. 3. Whether the professional involved in submitting the survey had any reflections on how EFH was recorded or how relocations were used in their service as a result of participation in the project.
Survey returns and transcribed interview data was initially analysed to report the rate of, and associated rationale for, relocations in cases of EFH in participating sites (Firmin, Wroe and Bernard, 2021). To achieve this, sites were organised into three groups -those whose survey returns evidenced that they relocated less than 5% of all young people they were supporting due to EFH in September 2019; those who relocated 5-10% of young people; and those who relocated over 10%. Interview data was organised according to these three groups, and through a process of thematic analysis (Braun and Clarke, 2006), patterns were identified in the rationale for relocating (or not) young people within each group. Whilst the rationale(s) provided were multi-layered and varied by group, the ways that professionals viewed (and worked with) relationships appeared a central consideration in the use of relocation. Indeed, it was evident that young people's 'relationships' were a central feature both in the decision-making surrounding relocations and in the nature of their use. This prompted the research team to conduct secondary analysis of the data collected during stage 1 interviews; with the goal of understanding the rationale for relocating young people affected by EFH through the lens of relationships and relational intervention.

Secondary Analysis
As Heaton (2008) has demonstrated, secondary analysis is a helpful methodology for offering new angles that are divergent from the original purpose of a research study. The approach allows for further scrutiny of themes that were -in the case of the original study -one element of the overall result. As noted above, in the first phase of the Securing Safety project, 'relationships', and concern for relational safety, emerged as a sub-theme in the rationale(s) provided by professionals for using relocation in response to EFH at the rates at which they did (the 'why'; it was used). Relationships also seemed to characterise the nature of relocation process itself (the 'how' it was used). To build the depth of understanding that this dimension warranted, secondary analysis was required.
Researchers undertook secondary analysis of interview transcripts from stage 1 of the Securing Safety project to explore: -What relationships (practitioner-young person, contextual, or parent-child) social work professionals prioritised when describing their use of relocation in response to EFH -How social work practitioners described using relationships prior to, during, and after relocations in cases of EFH -The extent to which participant accounts of relocations reflected, or challenged, wider research, practice, and policy developments in England that have promoted interventions with EFH via practitioner-young person, peer/community-young person relationships, or parent-young person relationships To achieve this, the data was analysed in three phases.
In the first phase, all original transcripts and the primary coding undertaken were reviewed. Queries were run in NVivo 11 on all child-nodes from the primary analysis which related to relationships or relational safety. Exported results were reviewed and irrelevant references discounted. The process generated 57 references to relationships across the 15 interviews, with at least one in each interview transcripts. Interview transcripts were reviewed on a final occasion to identify any references to relationships that may have been missed when the data was initially coded. No additional references were identified.
All 57 references were initially organised by relationship-type: those that related to young people's relationships with professionals, families, peers, and communities. This work surfaced findings on the types of relationships that participants referenced most when describing their use of relocation, and their rationale for avoiding or using this intervention.
All 57 references were then reviewed manually to identify themes in how interview participants described using (working with or providing) relationships prior to, during, and following relocations. Framework analysis (Ricthie et al., 2003) was 1 3 used to cluster the themes that emerged, and cross-check them against different types of relationships that were discussed (practitioner-young person, family-young person, peer/community-young person). In this process, similarities and differences between the themes identified for each set of relationships were considered, and some were collapsed. This process identified three dominant ways in which professional described using relationships in the context of relocations.
Finally, two researchers considered the results from the first two stages of analysis and mapped them against three relational responses to EFH detailed at the outset of this paper. The exercise helped us to establish the extent to which the narrative of our study participants aligned with, were divergent from, or challenged evidence related to relational responses to EFH; and provided us with a roadmap for synthesising relational practices and relocation interventions in the future.

Ethics
The Securing Safety project received approval by the ANON Ethics Panel, where the researchers were based at the time of the study. Later collaboration agreements between Securing Safety and Securing Safety supported onward use of the data for publication purposes by the research team until December 2022. Additional approval for the study was granted by the Association of Directors of Children's Services.

Relationships that Were Prioritised in Social Work Accounts of Relocation
All 16 interview participants commented on young people's relationships when talking about relocations in response to EFH. As Table 2 illustrates, this included young people's relationships with practitioners, parents/carers, their peers, and their communities. Young person-practitioner relationships were most frequently referenced during interviews and were raised by the greatest number of participants.

The Use of Relationships in Professional Accounts of Relocation
Three themes characterised how social work professionals described using relationships in the process of undertaking, or avoiding, EFH-related relocations. They used relationships to provide consistency, foster collaboration, and create safety, for young people affected by EFH.

Consistency Through Relationships
Professionals described efforts to maintain 'contact' between young people and practitioners that they knew (n = 14 references across n = 7 interviews), during relocations. For some participants, maintaining this type of contact was a matter of meeting statutory obligations of social work intervention: So they'll definitely have their lead social worker that stays with them, because they have statutory responsibilities (Practice Manager, Area J) However, professionals also described how sustained contact between a young person and practitioners they knew offered consistent relationships at a time of relational turbulence: where other relationships were fractured (they had been removed from their families and friends) or new (with foster carers or staff running residential homes).
… there's really good examples of working where social workers have gone up initially every week or every fortnight, they've worked really hard to support contact that's been supported by staff to build relationships (Strategic Lead, Area E) Where possible, we ask local services to continue to support young people for that kind of continued relationship when they're placed certainly in the region (Exploitation and Missing Coordinator, Area F) Comparatively, little attention was paid during interviews to the need for consistency in young people's relationships with their families, peers, and wider communities following a relocation. Two professionals described facilitating contact with families when asked to describe the interventions or support they offered to young people who had been relocated: We would be facilitating family contact and family time between children and families (Quality Assurance Service Manager, Area I) Two professionals also described efforts to maintain young people's contact with their peers during a period of relocation: We do also try and ensure, where possible, that contact is of high quality as well, so they maintain their relationship here. Depending on their peer group as well, you would hope that they'd be able to continue their friendships. (Lead Social Worker, Area K) Four professionals acknowledged that young people affected by EFH expressed a need for consistent community relationships; both in areas from which they had been moved and in areas where they had been placed. …in our experience, generally the children that we have placed out of county are like homing pigeons and will come back. (Principle Safeguarding Officer, Area H) … for us, we know these children aren't likely to come back to [AREA] and how we maintain connected networks or for some of our children who are out of area who want connected networks where they've been placed, how that's sustained or supported in the longer term. I think we need to do some more thinking on that really (Quality Assurance Service Manager, Area I) However, for the most part, they described struggling to meet young people's desires for relational consistency with communities and identified that more work was required in this regard: …what do you do about getting them back into the area, or is the longer term plan for them to settle where they are and then it's about how you do that and set up their support networks. (Assistant Director, Area B) …you've really got to think about how you're going to reintegrate them back into an environment that you've identified as being potentially unsafe because if you don't and they reach that 18th birthday, they're coming anyway without that period of readjustment back into the area. (Senior Manager, Area A) Rather than provide accounts of consistent relationships during relocations, three professionals framed relocations as a risk to consistent relationships between young people and practitioners. Two of these professionals also described consistency in familial relationships as a feature of effective responses to EFH. In these accounts, participants referenced relational consistency as a reason to avoid, rather than to use, relocations; to the point that some were working in services that had decided against moving young people affected by EFH: We probably about five or six years ago took the decision that actually placing children in external provision … isn't the best thing, because it takes them out of the area, it takes them away from their family, and you can't support them as well. So we've worked hard to bring children back into the authority. (Head of Child Placements, Area C)

Collaboration Through Relationships
Professionals identified relationships of collaboration as a core component of both successfully relocating, and decisions not to relocate, young people affected by EFH. They described collaborative relationships between young people and professionals, and between families and professionals -and to a lesser extent between all three groups. References to collaboration with peer groups/friendships or through community relationships were noticeably absent in the data.
Four professionals referenced collaborative relationships between young people and practitioners in their accounts of relocation. They described these 1 3 relationships as built on, producing, or cementing, trust. Trust, in turn, facilitated the creation of 'safety plans' at the end or outset of relocations; plans which young people agreed with -and in some cases had designed themselves: …that handover was done very closely, and with her as well, she was fully part of that, that was a family decision to move her back, the family back to that area, the police had quite a lot of reservations about that because of the risk that they thought she may be at, but she met with them and she developed her own safety plan, and I read it, it was amazing, they literally said, "Give us your safety plan and we'll support that where you need us to", so that was done, that's just an example of that on (Exploitation and Missing Coordinator, Area F) Collaborative relationships were presented as a tool for creating sustainable rather than temporal safety: facilitating the development of plans that outlived a young person's 18th birthday and/or the time limits of any statutory care plan. To achieve this outcome, professionals described collaborative relationships with young people as ones in which practitioners had to take, accept, or sit with, risks; rather than trying to control or eliminate them: We would try and use a safety plan that the child has developed themselves, so there might still be an element of risk involved, but that would be owned by them and they could then make decisions based on who they want to speak to, and we would be able to keep them safe, because we've found that if you try and cut certain people from someone's life, that sometimes elevates the risk rather than reduces it. (Lead Social Worker, Area K) Some professionals also described characteristics of practitioner-young person relationships in which collaboration flourished. They portrayed the practitioner role as being to support or enable young people impacted by relocations and/or EFH to take decisions/actions -rather than seeking to control those decisions: It's about just being there mainly, I think, more so than enforcing your presence on them. (Senior Manager, Transitional Safeguarding, Area A) Their accounts implied that for collaborative relationships to foster successful relocations (whether permanent or temporary), professionals/systems needed to handle the discomfort of risk and relinquish elements of control.
All examples that professionals offered of young person-practitioner collaboration occurred in relationships that were used during, or near the end of, relocations, whereas nearly all of professional descriptions of building collaborative relationships with parents/carers/families related to action taken prior to relocations (n = 7). Professionals explained how they used collaborative relationships to create safety plans in partnership with parents/carers. They also offered examples of interventions they used to strengthen parent/carer-young person relationships, thereby increasing a pull towards home, and away from any extra-familial contexts/relationships where they had been harmed: 1 3 [we use] a parenting worker that works with the parents and or foster carers that's looking at their ambitions (Practice Manager, Area J) Then we'd also try and ensure that a child can remain in the family home, especially where it is extra-familial risk, because we work to the knowledge that peers or the perpetrator are more influential, so we would always try and keep a child in their home.…So we could use family group conferences to try and address the risks within the family and use their strengths (Lead Social Worker, Area K) In this respect, collaborative relationships between parents/carers and practitioners were built to avoid relocations. For professionals who described the role of practitioner-parents/carer collaboration, relocations occurred when such relationships failed to provide safety for young people. One professional described work undertaken to develop a plan for a young person through a collaborative relationship between practitioners, parents/carers, and young people during (rather than prior to) a relocation. The approach appeared fruitful: And they've moved from positions of conflict to positions of agreement across the family network. They've also supported the network around the child having ownership of their network around them and legally some of the care planning (Strategic Lead, Area E) However, such examples were not characteristic of the wider dataset.

Creating Safety Through Relationships
The role of relationships in creating safety, and the extent to which safety through relationships could be created during relocations, emerged as a final theme in the dataset. In cases of EFH, young people are moved away from predominantly peer or community contexts in which they have experienced harm. Accordingly, a small number of interview participants described efforts to build safety through/in peer (n = 3) and community relationships (n = 2) prior to and during relocations: two young women and also another two that were friends with them, I was talking about, they did try to do joint group work with those young people, the base simply being perhaps if we could influence the dynamic between the children, we could increase safety within their own relationships (Senior Manager, Area E) …there should always be work done to try and move young people back to their homes, not necessarily back to their parents, but back to the area where they come from. (Quality Assurance Service Manager, Area I) In most instances, work to build safe relationships for young people within their communities or amongst their peers resulted from collaborative working between young people and practitioners. Professionals referenced how, if young people were going to live safely in the future -beyond the life of statutory intervention -they needed access to sustained protective relationships. Moreover, for young people to buy-in to plans, such plans needed to feature relationships that were important to them -including those that were extra-familial. However, this was not a dominant narrative across the interviews: three participants discussed safety through peer relationships and two reflected on building safety through community relationships.
In keeping with the theme of collaboration, one participant also explored how safety was created through the relationships forged between young people and practitioners -including those that ran settings where young people were placed. I was quite struck by the relationships that those places have built with the young person and the information sharing with the social worker, and I know for definite that at least one of those, the key worker in her placement is probably the only person she speaks to… (Exploitation and Missing Coordinator, Area F) As with wider professional accounts of relationships in the context of relocation, building safety through familial relationships was described as something that happened prior to, rather than during, relocations (n = 4). When efforts to build safety through familial relationships proved ineffective, relocations were used; and at this point, it appeared that safety through practitioner-young person relationships was prioritised, as well as some consideration of community and peer relationships.

Social Work Accounts of Relocation Through the Lens of EFH Interventions
As outlined earlier in this paper, relationships feature in three approaches to EFH intervention that have been promoted by England's policymakers and researchers: trusted relationships between young people and practitioners; building safety in young people's peer, school, and community relationships; and strengthening relationships between young people and their parents via relationship-based social work. To varying degrees, the responses offered by professionals in this study reflect these three approaches; they also indicate challenges that may arise when trying to hold all those relationships in mind during relocations of young people affected by EFH.
By prioritising descriptions of consistent practitioner-young-person relationships, professional accounts of using relationships in the context of relocations reflected research that promotes trusted relationships with practitioners as a key EFH intervention. This was reinforced by the examples of collaborative relationships in which 'trust' was positioned as central to building shared safety plans between young people and practitioners. It is notable that only one professional described safety being created in a practitioner-young person relationship. Practitioner-young person relationships are largely time-limited -bound by the length of interventions or the funded-life of commissioned services, whereas young people's relationships with families, peers, and wider communities were described as enduring and this may be why creating safety in these relationships was prioritised in professional accounts.
Despite the dominance of practitioner-young person relationships in professional accounts, references that were made to consistent contact, and collaboration with, young people's parents/carers were reflective of research that promotes relationshipbased practice. These accounts were particularly focussed on interventions to build safety in the family system prior to relocations; attempting to mitigate the pull/ influence of extra-familial relationships. However, the limited references made to responses that prioritised consistent contact with families during relocations illustrate how, once a relocation has occurred, professionals in this study struggled to maintain accounts demonstrative of relationships-based practice. Instead, efforts to build trusted relationships between young people and practitioners were prioritised in most professional accounts -rather than being given equal attention.
A much smaller number of interview participants described working with young people's extra-familial relationships prior to, during, or after relocations. When they did, professionals depicted tentative efforts to engage with peer relationships, and recognised a need to facilitate community contact, for safety to be sustained. However, the narrative that largely emerged across interviews was a lack of clarity about how consistent relationships, or collaboration, with peers and communities could be facilitated in the face of extra-familial risks. As such, whilst some professionals appeared to have ambitions to use relationships during relocations in ways that aligned to the emergent evidence on EFH interventions featuring extra-familial relationships, they were some way off making this a feature of routine practice.
As the above paragraphs suggest, interventions via trusted practitioner-young person relationships, peer/community relationships, or parent/carer relationships were discussed by participants in a siloed fashion. Relationship-based practice with parents/carers was dominant prior to relocations (via efforts at collaboration). Trusted relationships between practitioners and young people were drawn upon during, and at the end of, relocations (via consistent contact and collaboration on safety planning). Extra-familial relationships were noted as important, particularly for facilitating sustained safety beyond the life of relocations, but their use in EFH interventions was largely an ambition rather than a reality (by using relationships to create safety). As depicted in Fig. 2, professional accounts rarely described drawing together all three relational approaches to EFH intervention at the same time-point of relocation (prior to, during, or at the end); nor did they use these different relationships in the same way (to offer consistency, to collaborate, or to achieve safety).
Nonetheless, green shoots of practice representing an intersection between research on trusted practitioner-young person relationships and peer/community relationship were evident in a small number of professional accounts. In these examples, professionals described relocations where consistent relationships between practitioners and young people fostered opportunities to collaborate on safety plans,  and these safety plans featured ambitions for safe community and/or peer relationships. In these accounts, professionals recognised that plans to end relocations needed to support the development of safety in young people's relationships with peers and/or home communities: and through these relationships, safety could be built in contexts previously associated with EFH. Other professionals described ambitions to develop such an intersecting approach, where they were able to use practitioner, peer, and community, relationships to improve the efficacy of relocations as a response to EFH.
As such, professionals who participated in this study described needing to engage with three sets of relationships (practitioner, familial, and peer/community) when responding to EFH; and in the case of relocations, needing to do this in at least in two places: the home authority (Fig. 3) and the area where a young person is placed (Fig. 4). Figure 3 illustrates the dominant account in interviews that, during a relocation, a professionals' focus shifts away from collaborative relationships with parents/carers and towards consistent and collaborative relationships with young people. Yet in both pre and post relocation narratives, there seems to be confusion about what constitutes the young person's relational system -and the relationships in which safety can be realised and sustained.

Discussion and Conclusion
Professionals in this study described relationships that are both protected and disrupted when young people are relocated due to EFH. Evidence of relationshipbased practices with parents/carers, efforts to build trusted relationships between young people and practitioners, and to a lesser extent recognition of extra-familial relationships were all represented in their accounts. When taken together, these accounts illustrate a struggle to represent all three approaches to relational intervention in the practice of relocation; as well as an inability to achieve safety through these various relationships in the context of EFH. England's social work policy and practice system does not currently support the adoption of an integrated model of relationships-based practice in situations where relocation is used in response to EFH. Firstly, the individualised nature of case management approaches to social work often views young people abstracted from contexts. This has proved challenging for developing any form of relational work (Trevithick, 2014) and is therefore ill-fit to supporting professional responses to multiple, and potentially competing, relationships. England's child protection system is also more broadly built on the concept of parental responsibility -positioning state intervention at the point where action or inaction by parents poses a risk of harm (Featherstone et al., 2018). Professional accounts that depicted relational working with parents prior to, rather than during, relocations are somewhat illustrative of this parent-state relationship, where prior to relocations, social workers prioritised supporting parents/carers to fulfil their 'responsibility' to protect their children. This focus declined once the state intervened by relocating a young person away from their home communities.
Moreover, multiple sector-wide challenges constrain practitioners from working in the dynamic, multi-context, relationships-based way that responses to EFH may require, such as frequent organisational change; the proceduralised and overwhelming nature of social work; and a culture of public scrutiny and blame (Lonne et al., 2016). This is exacerbated by the significant levels of risks posed to young people by EFH that practitioners are asked to hold often without adequate organisational containment (Ruch, 2007). Whilst practitioners may be aware of and wish to work in a relational way, systemic factors can nevertheless lead to a focus on closing casing and drawing boundaries in order to cope (Lonne and Parton, 2014), and to practitioners adopting singularly focussed relational routes of intervention. And yet, as outlined in research detailed at the outset of this paper, and reinforced by professional accounts in this study, in cases of EFH, young people need help to manage a range of relationships -ones that become even more complex during relocations and that are not well-served in case-and-risk-management structures. For this to be realised, the organisational context needs to enhance conditions, which would enable complex, multi-context relationships be held concurrently.
There is also a need for further research that explores and communicates the possibilities of integrating various forms of relational intervention in cases of EFH. Firstly, each body of research that provides background for this paper asks practitioners, and the systems in which they work, to intervene through specific (and discrete) relationships as a means of safeguarding young people from EFH. This study suggests that all these relationships are important -and particularly so when relocations disrupt some and establish others. Echoing reflections from the work of Featherstone et al. (2018) to promote a Social Model for protecting children, the findings from this study indicate that responses need to '…recognise children within their communities and to work productively and collaboratively with a number of networks ' (2018:101).
Moreover, as extra-familial harm usually occurs during adolescence, it is likely that professional intervention will be time-limited, dependence of caregivers will be changing, and engagement in extra-familial relationships increasing. There is a role for researchers in offering more integrated conceptual and practice models of relational practice: in which young person-practitioner, young person-familial, and young person-extra-familial relationships are all accommodated, weighted, and valued. Further to this, given the significance of peer-relationships during adolescence, and their association to risk and protection in cases of EFH, it is critical that young people's peer groups are taken seriously as relational systems, and thought about alongside familial systems. Whilst practitioner preference for familial-based relationships is understandable, due to the focus within psychoanalytic and attachment theories on early relationships between infants and their caregivers, there is considerable scope for applying these beyond their original focus (e.g., Armstrong and Rustin, 2015) both conceptually and empirically, in order to expand psychosocial understandings of the particular nature of relationship-based practice with adolescents.
Finally, a study into specialist foster placements for young people who have been sexually exploited (Shuker, 2013) demonstrated that safety was multi-dimensional and needed to be engaged with as such. It found that when placing a young person in out-of-home care due to exploitation, often their physical safety was prioritised over and above the safety they realised through relationships (relational safety) as well as their internal well-being (psychological safety). This study thickens our understanding of relational safety in Shuker's model; pointing to the various relationships that practitioners protect or disrupt in the process of relocation. A dual but interrelated focus, therefore, is now required from research: firstly to expand theoretical 1 3 understandings of adolescents affected by EFH within multi-relational systems; and secondly to establish the organisational and practice conditions in which responses to EFH hold together and engage with the multi-layered relational needs of young people.

Limitations
The results draw on the views of 16 interview participants drawn from 14 local areas in England and Wales who participated in phase one of the Securing Safety project. Whilst the areas were mixed in terms of geography and demography, they are not a representative of sample of England and Wales. Moreover, the views of each professional interviewed do not necessarily reflect those held by others in their respective organisations. These limitations are managed by situating the dataset within a wider narrative on relationships, safeguarding interventions and extra-familial harm; and by presenting findings as how participants viewed relationships and relocation rather than how relationships are used/featured in all relocations.
Secondly, the issue of 'relationships and relocations' emerged as a sub-topic within a project that was concerned with the rate, cost, and impact of using relocations in response to EFH. As such, interview participants were not explicitly asked about relationships, and may have offered different answers in such a circumstance. Nonetheless, this study demonstrates what professionals said about relationships in the contexts of relocations when unprompted -and so provides a more organic, albeit less purposeful, account of practitioner perspectives.
Despite these limitations, the results of this study raise questions for policymakers and practitioners about how to situate relationships in the care plans of young people who are relocated due to abuse/harm beyond their families, and signal a need for an integrated evidence base on relational responses to EFH in the future.