“I’m not Just Some Criminal, I’m Actually a Person to Them Now”: The Importance of Child-Staff Therapeutic Relationships in the Children and Young People Secure Estate

Abstract

Therapeutic relationships between children and staff across the Children and Young People Secure Estate (CYPSE) is an under-researched area. The aim of this research is to explore the research question, ‘what are the key elements of the development and maintenance of effective therapeutic relationships?’ Qualitative semi-structured interviews and a focus group were conducted with children in Young Offender Institutions, Secure Training Centers and Secure Children’s Homes in England between February–November 2019 (N = 28 children; 25 interviews, 1 focus group). Framework analysis and grounded theory methodology were applied to the data, and 13 themes were generated. The facilitating properties of good communication, understanding, reciprocal respect, trust, and a sense of fairness in these relationships are highlighted. Positive relationships may be facilitated through the formulation process. The varied experiences of children are considered, highlighting the need for knowledge sharing and training staff in effective helping skills.

Keywords:

• Children and Young People Secure Estate
• therapeutic relationships
• formulations
• SECURE STAIRS
• trauma-informed care

The links between childhood adversity, trauma and high-risk high-harm behaviors are widely known (Campbell et al., 2016). This includes long-term correlates, and evidence of the cumulative effect of childhood trauma on the later likelihood of engaging in high-risk high-harm behaviors (Laceulle et al., 2019). In the global north, including Europe, US and Canada, the prevalence of challenges such as depression, low mood, learning difficulties and neuro-disability have been demonstrated as significantly higher in children and young people (hereafter ‘children’) accommodated by the youth justice system, compared to children in the general population (British Psychological Society, 2015; Hindley et al., 2017; MacDonald et al., 2013; Odgers et al., 2005; Pyle et al., 2016). Research suggests children accommodated by the youth justice system in the global north are also more likely than the general population to have multiple complex needs, including those that stem from high levels of victimization and vulnerabilities (Chitsabesan et al., 2006; Harrington et al., 2005; Lader et al., 2003; Pyle et al., 2016). They are more likely to engage in high-risk, high-harm behaviors, including substance misuse and harm to self and others (Department of Health, 2009; Harrington et al., 2005; MacDonald et al., 2013; Odgers et al., 2005; Pyle et al., 2016; Ryan & Tunnard, 2012). Further, school exclusions, poverty and being accommodated by a local authority (cared for by the state) are also key contextualizing factors which research suggests are more prevalent in this cohort (Her Majesty’s Prison & Probation Service, 2019; O’Neill, 2001; Pyle et al., 2016; Taylor, 2016). Contextual factors more likely to have been experienced by children in the youth justice system include childhood bereavement, childhood experiences of parental separation, domestic abuse (Vaswani, 2018) and experiences of gang involvement, sexual and criminal exploitation (Department for Education, 2016). These in turn are linked to insecure attachment (Baer & Martinez, 2006), impacting on opportunities to form stable relationships over time.

There is an ethical requirement of the accommodating systems to meet the high levels of complexity and need in this population. Further, where more disciplinary approaches are adopted, such as restraints and seclusion, there is a risk of re-traumatising children (Bryson et al., 2017; MacDonald et al., 2013). Therefore, there is a need for emotional support from staff across the youth justice system (Biggam & Power, 1997). In response, a new Framework for Integrated Care, termed SECURE STAIRS, is being implemented across the secure estate in England (NHS England, 2018; Taylor et al., 2018). The Children and Young People Secure Estate (CYPSE) in England consists of youth detention accommodation, comprising Young Offender Institutions (YOIs; run by HM Prison’s Service, these are settings that securely accommodate children aged between 15–18 years old, placed by the youth justice system), Secure Training Centers (STCs; run by private companies, these settings securely accommodate children aged between 12–17 years old, placed by the youth justice system) and Secure Children’s Homes (SCHs; run by local councils, these settings provide secure residential care for children aged between 10–17 years old).¹ The Framework is a whole-system approach to care, which involves training and supporting all staff, from care, custody, health and education teams, to provide more trauma-informed, developmentally attuned, psychologically-based care; underpinned by a multi-agency, co-produced formulation involving multiple stakeholders. The co-production of each child’s formulation, a psychologically-informed shared understanding of child’s circumstances and difficulties, is supported by regular multi-disciplinary meetings with the child, or child and family, or whomever has parental responsibility, as relevant. While informed by staff training, the structure of the meetings and the development of the formulation was locally defined (Taylor et al., 2018). This is an evolution of forensic formulations, which have been previously described as covert or implicit (Hart et al., 2011), and as also outlined by Harvey et al. (2015). A central part of the formulation meetings is hearing the voice of the child, enabling them to tell their story in their own words. The passing of memories through a narrative is a key element in people’s ability to understand themselves, to provide clarity for the present circumstances, using that understanding to set goals for the future (Rose, 2014). Responding with empathy and understanding to children’s stories is considered vital for staff in the CYPSE, to engender supportive relationships to help children (Brown et al., 2014; Rose, 2014).

Trauma-informed initiatives focus on efforts to understand individuals and the contextual factors contributing to their existing difficulties (see Center for Substance Abuse Treatment (US), 2014; Hodas, 2006) which can be conducted through this process of formulation meetings. The aim of trauma-informed practice within this context is to provide a safe, trusting, supportive environment for children to reduce the risks of re-traumatisation; to break the cycle of trauma and reoffending. One of the main critical factors in the management of children’s behavior and understanding their needs is their relationships with peers and staff (Ipsos MORI, 2012; Liebling et al., 1999). There may be existing, ingrained distrust by children of the system which seeks to help, based on prior adverse childhood experiences (Gormally & Deuchar, 2012; Howerton et al., 2007) potentially leading to further complexities and challenges within the CYPSE. While building trusting relationships has been viewed by staff as a key element in the engagement of children in interventions offered to them, some staff also report the key driver for building these relationships as being motivated by the need to manage children’s behavior (Gyateng et al., 2014). Further, the ability of secure units to effectively work with children in their care is considered to be contingent on staff’s ability to contain the immediate effects of their behavior, that is to intervene early and manage risk, then leading onto understanding the difficulties leading to the behaviors through a developmental lens (Rose, 2014). Prior research in the CYPSE has found that the majority (between 62% (YOIs) and 82% (STCs) across setting types) of children felt their relationship with staff was good (Gyateng et al., 2014). Under the Framework for Integrated Care (SECURE STAIRS), relationships within the CYPSE are proposed as a key driver of change for children (Taylor et al., 2018). Child-staff relationships in the CYPSE are important to reduce anxiety, depression, and hopelessness (Biggam & Power, 1997). A study of children in the juvenile justice system in the United States used a youth mentoring relationship quality inventory, to examine the ‘most helpful’ (as determined by children) relationship with a care worker. Children with relatively high positive scores across the three relationship quality subscales of satisfaction, closeness, and coping, reported the greatest perceived likelihood of success on release in terms of social networks, substance abuse/reoffending, and conflict reduction (Marsh & Evans, 2009). Further, children have highlighted the importance of being treated with respect, taken seriously by staff and for staff to be people they can respect and rely on. They also expressed the value of consistent, reliable adult role models in their lives (Lyon et al., 2000). Although in a community-based rather than institutional setting, much can be learnt about the role of mentoring relationships between children and nonparental adults. Research exploring the effects of mentorship on at-risk children suggests the role of a trusted nonparental adult mentor is pivotal (Beam et al., 2002). Mentors can be defined as a responsible adult that a child feels able to trust and believes is willing to help and treat them with respect. These characteristics may also be important for positive child-staff relationships in an institutional context and could be a protective factor against high-risk behaviors. For example, strong positive associations have also been demonstrated between adolescents having an adult mentor and decreased participation in high-risk behaviors, including weapon carrying and drug and alcohol use (Beier et al., 2000). Additionally, the Early Intervention Foundation Review (Lewing et al., 2018) into the role of trusted relationships for children vulnerable to sexual exploitation and abuse concludes that the existence of trusted adult-child relationships is associated with better outcomes for vulnerable children and with reduced risk that children who have experienced maltreatment becomes perpetrators themselves. They note evidence that children who have experienced abuse may find it harder to build trusting relationships with adults but that positive child-staff relationships are key elements of helping approaches and are associated with positive outcomes.

The aim of this research is to fill a research gap, exploring child-staff therapeutic relationships in the CYPSE in England. The research question is, ‘what are the key elements of the development and maintenance of effective therapeutic relationships in the CYPSE?’ ‘Effective’ in this context relates to the impact on children’s experience of the CYPSE, their sense of self and future outlook, including mental health, wellbeing and likelihood of reoffending.

Materials and methods

The APA journal article reporting standards for qualitative research guidelines (Levitt et al., 2018) were followed.

Setting

Participants were involved in a wider evaluation of a Framework for Integrated Care programme (SECURE STAIRS), commissioned by NHS England and NHS Improvement across the Children and Young People Secure Estate (CYPSE) in England (D’Souza et al., 2021). The Framework for Integrated Care programme (SECURE STAIRS) provides a new whole systems way of working in the CYPSE that involves training staff to provide more developmentally-attuned, psychologically-informed care, which is centered on comprehensive, co-produced assessments of young people’s needs to ensure that all needs are identified (Taylor et al., 2018). At the time of analysis, the evaluation was at a mid-point in time, so sites accommodating the participants were at different stages of implementing the Framework.

Participants

Participants were recruited from five sites. Children who participated were accommodated by SCHs (N = 9), STCs (N = 4) and under-18 YOIs (N = 15). Within these settings, all children over 16 years who were able to provide consent were eligible to take part in the interviews. Children aged 16 or over were identified by staff to participate in a process of convenience sampling. Children under 16 years were not included due to the limitations of obtaining parental consent. The study and interview process were introduced to the children and children who expressed an interest in taking part were subsequently inducted into the study by a member of the Research Team (SD, RL), where they were offered the opportunity to ask questions prior to participation.

Ethical approval for the wider evaluation and associated research, was granted by the London–South East Research Ethics Committee and the Health Research Authority for the data collection with children (IRAS project ID: 242383; REC reference: 18/LO/1569). Her Majesty’s Prison and Probation Service granted ethical approval for data collection with children (2018-274). Informed consent was obtained from all participants. Identifiable details were deidentified at the point of transcription.

Data collection and analysis

Qualitative semi-structured interviews and a focus group were conducted with children between February-November 2019 (N = 28 children from 25 interviews and 1 focus group). The interviews and focus group were audio recorded; data files were securely handled and transcribed ‘smart verbatim’. A small number of interviews (N = 6) were recorded using researcher field notes due to audio recording limitations at sites. Twenty-four were conducted face to face and four were conducted via telephone. Interviews were held for between 5 minutes 47 seconds and 41 minutes 21 seconds, with an average interview time of 17 minutes (SD = 10.57). The interview topic guides were created by the Research Team and Steering Group, for the overall evaluation. Topic guide questions included:

What has been your experience of this service?; Do you think staff here have a good understanding of your views and experiences?; Do you feel staff listen to you?; Do you feel staff take your views and worries seriously?; Do you know who to contact when you need to talk to someone?; Do you think different staff know your story?; Have you noticed any changes in how staff and young people interact since you’ve been here?

To gather views on the formulations process directly, the process was briefly outlined to the participants, and they were asked if they had been involved, and what their experiences of that were. The sample size was determined to achieve diversity in views and perspectives so that we were confident that we were able to reflect similarities and differences across a range of experiences (Morse, 2000). All data analysis was conducted in the Nvivo 11 qualitative software package (QSR International, 2020).

The first stage of analysis comprised a Framework Method (Gale et al., 2013) where researchers and clinicians in the Evaluation Team and Steering Group collaboratively derived an analytical framework for the overarching evaluation. The Framework Method is suited to research conducted by a team, rather than individual researchers and is appropriate for use when a large amount of data have been amassed and the study has specific research questions (Midgley et al., 2017). We followed the first four stages of framework analysis (transcription; familiarization; coding; and developing a working framework; Gale et al., 2013) to explore patterns in the data as part of the evaluation of the Framework for Integrated Care (SECURE STAIRS). This first step was principally a data reduction step, to understand the data at a high level for evaluation purposes, and to create a more manageable dataset to take forward into the next stages of analysis and interpretation.

The second stage of analysis followed a grounded theory methodology. The data organized within the Framework was revisited, each category in turn, and themes were derived through the analysis and organization of the transcripts, in a process of open coding (Corbin & Strauss, 2008). It was important to use an inductive method to explore the data at this stage, and to combine it with the deductive prior step. This allows the research findings to be grounded in the overarching methodology in the first instance, but then allows a broader analysis and understanding of the data to be explored through the development of secondary themes.

The Framework categories ‘Culture change—relationships with staff’ and ‘Formulations’ were the focus of the present analysis. For the third stage of data analysis, the data previously coded into these Framework categories was further coded into subsequent themes through the Constant Comparative Method (Boeije, 2002). The constant comparative method is an inductive analysis technique, which involves categorizing and comparing data, with the main aims of discerning conceptual similarities, to refine categories and to discover patterns (Tesch, 1990). The first author (JJ) coded all transcripts and the third author (RL) coded 25% of data. Areas of disagreement were subsequently discussed and aligned. Both coders had been previously trained in qualitative data analysis to Master’s (SD) and Doctoral (JJ) levels and were experienced in coding qualitative data. The themes were subsequently shared with the rest of the research team for review. The interviews/focus group analyzed in this study was reported in the final evaluation report submitted to the funders, but the analysis pertaining to the research question outlined in this manuscript was unique to this study and has not been reported elsewhere.

Methodological Reflexivity

The epistemological positioning of the coders (JJ, RL) and all researchers, prior knowledge and exposure to psychological theories, trauma-driven care and the political and structural context of the secure estate will impact the analysis and narratives in this research. The coders were mindful of these potential tensions, and during the analysis, made conscious attempts to remain open minded to deriving data driven results, within the context of the wider Realist Evaluation and the overarching framework model which in itself will impact the findings. The implementation of a Framework Approach meant the analyses were driven by theory such that a solely data driven approach may have derived entirely different results. One coder (RL) also conducted some of the interviews, which may impact the interpretation of the data presented, which may have been different, even subtly, to the interpretations made by the other coder. Any differences in understanding and interpretation between the coders were discussed and agreements were established. This convergence implies enhanced objectivity and accuracy within the Realist Framework (Madill et al., 2000). The varied experiences of the researchers are beneficial to the narrative and understanding of the data; more specifically, some researchers have research backgrounds, others have clinical backgrounds, however, none of the researchers have lived experience of being accommodated by the secure estate.

Further, the Research Team’s interactions with participants might have been influenced by their own experiences and prior assumptions. All data collection was conducted by two researchers (RL, SD). Topic guides were developed by the Research Team and Steering Group as part of the evaluation project, but the experiences and prior views of the interviewing researchers may have been expressed in implicit emphases on questions. Further, a prior distrust in the ‘system’ and authority, linked to adverse childhood experiences (Howerton et al., 2007), is likely amongst children in the CYPSE. Therefore, if participants perceived the researchers as professionals from within ‘the system’, it would potentially limit the candor of their responses.

Results

Participants expressed perceptions of staff members’ ability to help them, sometimes providing concrete examples of how they had been in receipt of helping behavior: “[…] They chat to you whenever you need it. […] that member of staff went out of his way to sort that out” (Child 2). There was a sense that staff making time for supporting them was important: “that’s also important for young people to have as well; staff that will make time for them and make sure they’re okay” (Child 4). Negative examples were also discussed, generally: “not staff that help you, you have to help yourself” (Child 6). Further, a sense of helplessness was also discussed, where, despite actors in the system being willing to help, the experiences of participants had not changed: “CuSP officers ask but they do nothing about it. They ask what they could do to help but it’s pointless” (Child 7). Specifically relating to formulations, participants provided explicit examples of how they linked the process to helping behaviors: “[…] for me it releases quite a lot of stuff that I’ve built up inside over the years, and it kind of eases that a bit because it’s out in the open. And staff can look at that whenever they need to look at that […] I think that helps” (Child 4). Other participants expressed difficulty in engaging with the formulation process, but that the professionals involved were able to work in a way that felt comfortable to the child “I struggle to […] open up about my story […] they were quite respectful, and they appreciated that it is always going to be difficult for a young person to share their story with people they don’t know […] they don’t push it” (Child 8).

Participants discussed inconsistencies/instability within the settings, leading to challenges with forming positive relationships with staff: “Some staff de-escalate a situation, but others can make it worse” (Child 16). Sometimes either poor relationships or the lack of positive, meaningful (facilitative), relationships with staff were attributed to high staff turnover, and therefore no opportunity to spend the time needed to build relationships: “there’s a load of new staff coming in and it’s building relationships all over again” (Child 4). In other instances, this was related to a reticence to open up and talk to staff: “I don’t really open up, I don’t really talk about my stuff” (Child 9). Further, participants discussed not being able to have facilitative relationships with staff for reasons which are unclear, which may be aligned to personal characteristics or other elements of relationship building: “you will have other staff you just don’t want to be around” (Child 10). Other elements of discussion contributing to the Lack of Facilitative Relationships theme were associated with staff not being invested in their roles: “don’t know how they [staff] passed interviews […] just do it to get their pay cheques” (Child 6).

Building facilitative relationships was described as helpful. This theme is the positive counterpart to the previous theme, evidencing mixed and sometimes conflicting views. Participants described their positive relationships with staff in terms of shared activities: “[…] chill with staff, play a bit of pool with staff” (Child 2). At times this was explained as being contingent on the behavior of the child and mutual respect: “You’ll be good for them, they’ll be good for you” (Child 2). Positive relationships seemed to be perceived as contingent on behavioral compliance. There was a general sense of getting on with all staff, or a particular staff member: “I just get on with any staff. I think most of the kids do. Literally everyone does” (Child 2) and “[…] there’s certain officers that I have more of a relationship with, and help more” (Child 3). Participants mainly reported having positive relationships with particular staff members, rather than in a generalized way.

Participants had a sense that staff cannot help, saying staff do not have either the ability or resources to help them, or that helping is not part of their roles. The reasons for this were not specific, but relate to a general sense of inaction: “I don’t think talking to staff changes anything” (Child 14). In this context the term ‘staff’ is used broadly and often referred, but was not limited, to frontline operational staff, educational staff and caseworkers. Due to the whole systems approach of the Framework for integrated care (SECURE STAIRS), children were not asked about specific staff roles, rather about their relationships with staff on the whole. There was also the perception that talking about difficulties was not a helpful strategy in itself: “Caseworkers will talk about it, not help you, but talk about it. There’s nothing she can help” (Child 14).

Participants described communication as a conduit to positive relationships. The vast majority of the examples was positive, relating to participants feeling able to talk to staff: “I’d talk to my case worker […] I can sit down and talk to her about anything” (Child 12) and “any staff member is there to listen” (Child 15). There was some indication that communication from staff in some YOIs was less nurturing compared to others where a more direct communication style is adopted: “the way that the govs speak to you is different […] in other YOIs I’ve heard it’s just like […] they open the door, they tell you what to do” (Child 13). There was also some evidence of fewer open communication channels and a sense that there were limitations to talking to staff: “if you speak to a staff member, they, they do pass it on, But that’s really all they can do” (Child 15). Participants expressed a lack of communication around the formulation meetings, where they stated that they had not heard about them or were not familiar with the concept.

When discussing staff’s capacity to understand children and/or to express empathy toward them, most participants said that they felt at least one staff member understood them. For example: “me and him have always been on good terms, understand each other and where we’re coming from” (Child 9). Participants referred to staff’s ability to understand them as the crux of the positive, or negative relationship. Time was also described as a facilitator and there was some suggestion that there had been improvements over time: “before I came here, I was nothing but horrible to people […] now it’s completely different. I respect people […]. The reason for that is the staff. They actually take their time in getting to know the young people in here now” (Child 8). This may be related to a shift in culture at the setting. There was a suggestion that staff were the agents of change for children. Participants provided detail on what that looks like in practice: “They’ve got this way of breaking things down, like your emotions and things. […] So, you actually go into depth on how to sort things out” (Child 4). Some participants’ reflections on them were negative: “They don’t know your story […] they just see you as that character” and “You talk to them about something then you hear them joking about it” (Child 10). Concerns were also raised that staff were racist and discriminatory toward children of color. They stated that some White frontline YOI staff, treated them with less respect than White children in comparable situations, for example, when waiting to hear an answer to a query.

Participants discussed a sense that staff are caring toward them. Most examples provided were positive: “Most staff […] they don’t care what you’re in here for, they care about you’re doing now to improve your life when you get out” (Child 17). As with the good staff theme, there was some indication, that some, but not all, staff displayed caring attributes: “The ones that I do talk to seem like they care. The rest don’t want to help” (Child 16). Where participants had been in the CYPSE for some time, they observed a positive change regarding caring staff. However, there was a sense that caring was not assumed, and an understanding of the child is a prerequisite to accessing caring relationships: “a lot of the govs […] they didn’t really know me, so, you know what I mean, they didn’t really care” (Child 18). The unique and distinct role of staff while also being caring and friendly was also discussed: “’Cause they’re not your mate, they look after you” (Child 19).

There was an overriding sense that relationships worked on a quid pro quo basis, where respect from staff is earned through children’s actions: “It’s all about how you approach them […] as long as you give them respect, they’ll give it you” (Child 9). The importance of respect and this being a two-way process was highlighted: “[…] people can have a laugh with staff, staff tell you like their life, you tell them yours, […] it’s all about the respect” (Child 15). There was also an indication of broken relationships, specifically related to participants not feeling respected. However, it is not clear whose respect this refers to: “[Staff assault] Only thing I can do to be respected, what else can I do?” (Child 6).

Most participants said they were central to the formulation conversations, that they felt listened to: “Formulation is based around the young person so you are heavily involved” (Child 20) and “it’s good when staff can hear your perspective” (Child 21). There was a sense that this engendered a facilitative relationship or change: “Every professional in that room listened to me […] they were interested in my feelings […] rather than just telling me how I am, how I behave and how they want me to change it” (Child 8). However, some participants expressed that they had not been involved in these processes and staff held the formulation meetings without them.

The sense of ‘good’ staff was expressed through words such as decent, good, brilliant, lovely and friendly. There was a sense that there are certain staff members who are considered ‘good’: “you get decent staff every day” (Child 2). In these examples, there is some suggestion that not all staff are ‘good’, but within this broad categorization, the differences are not apparent.

Most participants referred to trust as a lynchpin: “building up that trusting relationship. To the point where […] the first people they want to talk to is the staff because of the relationship they have […]. If there’s no trust, then the young people can’t get any help for any of their behaviors” (Child 8). For most participants, there was a sense that in general there was no trust between children and staff: “I think the trust isn’t really there” (Child 22). However, some participants referred to being able to trust some staff, but not all. Participants expressed that formulations in particular were a key facilitator in them feeling able to trust staff.

Participants expressed perceptions that staff either view and/or treat them like criminals and there was no consideration of the individual aside from this one-dimensional conception, for example, “[we are] all seen as prisoners, all in the same category” (Child 6). Specific examples of how these perceptions are explicit from staff were given: “[other young people] don’t want to be here because they feel like they’re getting treated like a criminal” (Child 17) and “one of the staff members […] said to one of the lasses that were on welfare, ‘the next place for you after here is […] you’re going straight to prison’” (Child 17). Participants provided further support for formulations being the key facilitator of reaching mutual understanding, relationships and ultimate change, for example when asked what it was that helped staff humanize: “it was that formulation meeting […] it tells them a bit more about me […] so I’m not just some criminal, I’m actually a person to them now” (Child 18).

There were varied perspectives related to a sense of fair treatment from staff, where some participants described general positivity within the context of day-to-day activities. Other participants felt staff treated them less fairly and, in some areas, the care provided was punishment-led: “sometimes they don’t maybe see what is really going on. I think they need to stop being so consequence, consequence, consequence […] some people can be a bit harsh” (Child 24). There was also an indication of a mixed picture within settings; and a sense of some behavior being acceptable for staff but not for children: “some of the staff […] like joke and like laugh around but they […] some of them do take it like, way too far. Which like, we’ll then kick off at them about it and then we get in trouble for it” (Child 25).

Strengths and limitations

The convenience sampling method meant researchers were reliant upon staff at the sites to identify participants. While all children over 16 years old who were able to provide consent were eligible to take part, staff may have inadvertently, or intentionally, acted as gatekeepers, and made decisions about who to invite based on their own assumptions and priorities. Any gatekeeping was not known to the researchers. Subsequently, the research may be subject to participation bias. The findings may be led by more articulate or forthcoming participants, while voices of quieter participants, or participants more willing to engage with people perceived as professionals may be unrepresented. Only participants over the age of 16 were interviewed, resulting in a gap relating to the voice of younger children. Additionally, we were unable to explore associations between participant characteristics and findings, and due to the risk of reidentification, we were unable to explore the impact of different staff roles on therapeutic relationships, leaving gaps in research prevailing. We also cannot be sure the views gathered are representative of the wider population within the CYPSE, nor whether they were specific to the different settings and the associated environmental factors, though efforts were made to recruit as diverse a sample as possible.

Due to security restrictions on audio recording, a small number of interviews were recorded using field notes. While this is a useful method for researchers, it is subject to potential memory recall bias, researcher bias and the context of the notes may be misinterpreted. However, the researchers ensured that accurate notes were taken.

Discussion

Overall, mixed experiences were represented in the findings. Participants generally discussed the importance of building facilitative relationships, in support of previous literature (Gyateng et al., 2014; Rose, 2014; Taylor et al., 2018). However, some participants described relationships as almost transactional, particularly expressed in regard to respect. There was a sense that some children knew how behave to get by, and there was an idea that staff treatment of children is contingent on how they behave. Where relationships are felt to be transactional there is a risk that children who have previously experienced grooming as a prelude to abuse or exploitation may perceive some similarity between aspects of their relationships with staff and with their abusers, impacting negatively on trust. A trauma-informed approach would therefore help children to perceive the differences in staff-child relationship, even when behavior management strategies (and therefore a degree of contingency) are required. In this context, that would mean staff being aware of the potential impact on current relationships of previous abusive or exploitative relationships and to adapt accordingly. While it makes sense that behavior is responded to contingently, there should be aspects of the staff-child relationship that are unconditional; staff-child trusted relationships are characterized by persistent care and compassion (Lewing et al., 2018). This may link to prior research where staff reported the key driver for building these relationships with children was motivated by the need to manage their behavior (Gyateng et al., 2014). Some participants described being understood by staff as a requirement to access caring relationships. Previous research in the CYPSE and residential care also found that children describe differences between staff who care about them and staff who do not (Bateman et al., 2013; Day et al., 2020; Moore et al., 2018). The sense of the limited ability of staff to help may be further linked to insecure attachment confounded by the power imbalances in the settings, and perhaps to a restrictive organizational structure. Reciprocity within therapeutic relationships with appropriate boundaries has been explored in previous research, where patterned and repetitive interactions between professionals and patients in a range of mental health settings have been demonstrated as promoting positive outcomes (Sandhu et al., 2015).

The present findings support prior research, where participants expressed a need to be treated with respect and to be taken seriously by staff and for staff to be respectful and reliable (Day et al., 2020; Lyon et al., 2000). Participants discussed listening skills and being “caring,” which may be personality attributes, but are also skills acquired through effective training and the trauma-informed systems of support provided through the Framework for Integrated Care (SECURE STAIRS). This may be evident in the varied experiences, where some settings were further along in the roll-out of the Framework for Integrated Care (SECURE STAIRS) than others at the time of data collection, which was explored in the wider evaluation. Prior research with young people in inpatient mental health and residential settings has highlighted the importance of staff being reliable and consistent, which in turn helps young people to feel valued, cared for and safe (Hartley et al., 2022; Moore et al., 2018). The present findings also align with previous research in mental health inpatient settings, where similarly, patients identified facilitative “traits” of staff including being understanding, accepting, trustworthy, friendly, and kind (Moreno-Poyato et al., 2016; Sweeney et al., 2014). Additionally, vulnerability, dehumanization and frustration were the main negative feelings identified by patients as barriers to positive therapeutic relationships, as well as a lack of time and unpredictability of the setting (Moore et al., 2018; Moreno-Poyato et al., 2016). The extent of the loss of liberty and autonomy amongst patients in inpatient settings has also been identified as a barrier to positive therapeutic relationships (Sweeney et al., 2014). This is a particularly pertinent, transferable finding to the present research. It is important to consider these synergies across settings, and to explore the impact of the pooled research findings on practice.

Children’s experiences of their relationships with staff across the CYPSE have been shown to differ depending on setting in previous research. For example, more children in STCs and SCHs rated the relationships as ‘good’ compared to those in YOIs, and there were significant differences in ratings across the YOI sites (Gyateng et al., 2014). This suggests that there are setting-specific variables that influence the quality of relationships that it was not possible to explore in the current research. Future research should further explore these differences through qualitative enquiry. Future research should also seek to explore the intersectionality between prior trauma and the ability of children to form meaningful, facilitative relationships within the CYPSE, this includes the staff perspective. Future research should also consider demographic characteristics: race, ethnicity, gender and specific complexity factors discussed here. Of priority is an exploration of the disproportionate number of children of color in youth justice settings (Lammy, 2017).

Implications

There were, some participants who expressed a lack of facilitative relationships; this may be due to the child’s general stance, or may be due to a disconnect with staff, or high staff turnover affecting the ability to form these relationships, which take time to form. High staff turnover is demonstrated as common in similar settings worldwide (e.g., Legislative Council Legal and Social Issues Committee (Vic), 2018). High staff turnover has been linked to high staff burnout and low satisfaction (Sheppard et al., 2022) and has been found to be detrimental to children’s sense of being treated as an individual in the CYPSE and in residential care (Day et al., 2020; Moore et al., 2018), suggesting an urgent need for exploration and cross-learning in this area.

It was suggested that as a facilitator for change, caring and warmth could be derived from the formulations approach, further evidenced by participants who discussed still having reservations about trusting staff, but that formulation had perhaps inched them closer to a relationship that looked more like trust. Building trust through relationships in inpatient mental health settings has been identified as a key facilitator to young people feeling able to talk to professionals. Further synergies between the current and this previous research relate to the importance of young people being involved in discussions about their care, and the stigmatizing nature of being labeled (Bjønness et al., 2020). Indeed, young people have described being involved in discussions from the onset of their inpatient stay made them feel safe to speak up amongst professionals, while being met with authority which was perceived as judgment, aligned with a loss of control that may trigger feelings of resistance to be involved (Bjønness et al., 2020). It is important that this learning is taken forward into work on formulations with children in the CYPSE, inpatient settings and community settings.

The importance of being understood, or for staff to express empathy toward children, is linked to communication; the experience of being understood, facilitated by the process of formulation, appeared to aid the establishment of trust in relationships, in support of previous research (Brown et al., 2014; Rose, 2014). This gives support to the emphasis in the Framework for Integrated Care (SECURE STAIRS) of shared formulations across staffing groups, rather than formulations held by a single clinician from the health team, as may previously have been the case. Formulations leading to the development of trust in relationships are in keeping with the theory that the experience of being understood (mentalized) and being recognized by another as an agentive individual triggers “epistemic trust”—the mechanism by which we learn from others (Csibra & Gergely, 2009; Fonagy & Allison, 2014). A question for further research is whether children can generalize help-seeking to future relationships. That trust is required for children to seek help from staff adds weight to the importance of starting to build relationships in their home area before they leave the secure setting. There is risk in not doing so, given previous evidence toward the preventive and stress-buffering impact of trusted relationships, in terms of vulnerability to abuse/exploitation and offending behavior (Lewing et al., 2018). Our title quote demonstrates the humanizing effect of sharing stories through formulations.

On an individual level, distrust may be an artifact of an insecure attachment to parental figures, particularly given that children accommodated by the CYPSE have often also been previously accommodated by a local authority (Her Majesty’s Prison & Probation Service, 2019). Further, being accommodated by the CYPSE in itself disrupts the usual course of children forming and developing attachments with family, education and support services (Day et al., 2020; Taylor, 2016) further confounding this difficulty, and these children are also at higher risk of experiencing prior adverse childhood events (Department for Education, 2016; Vaswani, 2018). Where children have experienced such abuses of trust and have perceived services as unhelpful, the development of distrust in others may be a reasonable—even adaptive—response. Indeed, such experiences are associated with insecure attachment (Baer & Martinez, 2006) and the development of epistemic mistrust or hypervigilance—in which one is unable to trust in information and help from others (Fonagy & Allison, 2014). This is particularly crucial for children in the CYPSE, who may not have access to supportive relationships outside of the setting (see Day et al., 2020). Further, the distrust of staff in the CYPSE may not solely be due to insecure attachments, but also from being previously failed by professionals in the care, education, and youth justice sectors. There is an assumption that children want to, or even should, trust adults which needs to be reflected upon, particularly when working with such vulnerable populations who are likely to have prior contact with a range of professionals, with some negative experiences. The development of a trusting relationship is suggested as a conduit for mental health patients to discuss their experiences with professionals (Battaglia et al., 2003). This is a key consideration when implementing formulations that include young people in the discussions.

Professional boundaries were alluded to in the present findings, where participants discussed some staff taking jokes too far, evoking a negative response from children which they then reprimanded. Power imbalances and fair treatment were further evidenced through quotes specifically addressing the nature of relationships, which could be interpreted as children having a sense of the different functions of various relationships, and what a helping relationship might look like. Histories of abuse and exploitation may leave children understandably more vigilant for signs they are being humiliated or taken advantage of Fonagy and Allison (2014). Remembering this would help to make sense of why the child “kicks off” at staff; perhaps as an understandable response, rather than something to be admonished.

The availability of mental health support and promotion across secure settings has been demonstrated as variable and fragmented across Europe, often delivered by external agencies (MacDonald et al., 2013). Our findings speak to a need for staff within these settings to develop and maintain meaningful therapeutic relationship with children in their care. The impact on the participants, of both the facilitative nature of positive relationships, and the detrimental nature of a lack of this type of support are evidenced in the present study. A shortage of staff and of funding have been identified as limiting factors to the implementation of health support to children accommodated by the secure estate, with a call for more investment, and the development of National and EU standards (MacDonald et al., 2013), with learning potentially applied further afield. Our findings could support this: through providing clear signposting to factors that underpin effective therapeutic relationships, as well as hindering factors that require urgent attention to underpin and support the development of such relationships, including the widespread challenges of staff retention.

Conclusions

Child-staff therapeutic relationships in the CYPSE is an under-researched area. This research goes some way to explore children’s experiences. Our research highlights the facilitating properties of good communication, understanding, reciprocal respect, trust, and a sense of fairness and understanding difficulties leading to challenging behaviors through a developmental lens. The varied experiences of children have been considered, highlighting the need for knowledge sharing and staff training in effective helping skills. Future directions should involve the inclusion of children and their families/carers as relevant, in all elements of forensic youth care, implemented in policy, with cross-learning applied internationally (Souverein et al., 2019).

Conflict of interest

No authors have declared any competing interests.

Acknowledgments

The authors would like to thank all the participants who took part, for their valuable contributions.

Data availability statement

The data that supports the findings of this study are available from the research team ([email protected]), upon reasonable request and approval from NHS England and NHS Improvement.

Additional information

Funding

This work is supported by NHS England and NHS Improvement and is sponsored by University College London (UCL). No grant number is available.

Notes

1 While some SCHs accommodate children placed for both justice and welfare reasons, others only accommodate children placed by local authorities under a Secure Welfare Order under the Children Act, 1989 for the protection of themselves and/or others (NHS England, 2018).