The views and experiences of suicidal children and young people of mental health support services: A meta-ethnography

Background: Suicide is amongst the leading causes of death in young people globally and a health priority worldwide. For children and young people (CYP) attempting or considering suicide there is no agreed treatment model. Development of treatment models should be informed by the views and experiences of CYP using services. Methods: Meta-ethnography was used to systematically identify and synthesise studies reporting the views of CYP who used mental health services following suicidal behaviour. Relevant studies were quality appraised. First order (participants) and second order (original author) data were translated to identify common and disconfirming themes and concepts. Translated findings were synthesised and led to a new hypothesis supported by additional ‘linguistic analysis’ of texts to construct a novel third

order line-of-argument. Results: Four studies conducted since 2006 in three countries involving 44 young people aged 11-24 years were synthesised. Translation revealed that suicidal CYP do not know where or how to access help, they cannot access help directly and when seen by mental health practitioners they do not feel listened to. Lineof-argument synthesis identified a silence around suicidality within the conversations CYP have with mental health practitioners and within academic research reporting. Use of the term 'self-harm' to encompass suicidal behaviours potentially contributes to this silence by avoiding the word 'suicide'. Conclusions: CYP who are suicidal need to have easy access to mental health services. When using services, they want to feel listened to and have suicidal feelings acknowledged. This involves professionals referring explicitly to suicide not just self-harm.

Word count: 250
Introduction Suicide is one of the main causes of death in young people globally (World Health Organization, 2014) and in some countries and communities it is the leading cause of death (Appleby, et al. 2017;Ljunggren & Nickel, 2016). Reducing suicide rates in children and young people (CYP) is a major international health concern and suicide prevention is a priority for many governments worldwide (Australian Government Department of Health and Ageing, 2008; HM Government Department of Health, 2015;McLean, Maxwell, Platt, Harris, & Jepson, 2008;National Office for Suicide Prevention, 2015;New Zealand Associate Minister of Health, 2006;Scottish Government, 2013;WHO, 2017). However, there are currently no agreed treatment models and the existing evidence base for managing CYP attempting or considering suicide is limited (Hawton, et al. 2015). The support and clinical interventions delivered to this population varies locally, nationally and internationally (Hawton, Saunders, & O'Connor, 2012). Risk assessment tools for suicide have been found to offer little predictive value, meaning that the criteria for determining who is considered suicidal and when suicidal behavior warrants therapeutic intervention is subject to individual clinical judgement (Carter et al. 2017;Chan et al. 2016). It is also acknowledged that many CYP who are suicidal or self-harming do not access support services (Taylor, Hawton, Fortune, & Kapur, 2009). For example, a confidential enquiry in England found that 62% of CYP under 18yrs who had died by suicide between 2014-2015, had  There has been a broad move towards service user participation and consultation in health research (Patrick, Guyatt, & Acquadro, 2017;NHIR, 2014). Additionally, legislation regarding the rights of children to express their views, be listened to, and be involved in decisions made about them, has indicated that health services should ask CYP about their views (United Nations Convention of the Rights of the Child, 2017). These insights could then inform the development and delivery of health services and treatment models that respond to the needs, wants and preferences of CYP (Facey, et al. 2010). In the last decade, there has been an increase in qualitative research concerned with exploring the views and experiences of children as patients in their own right (Kirk, 2007). This increase has also contributed to the growth in qualitative evidence synthesis where findings of individual studies are brought together to produce a new interpretation. It was the authors' belief that a body of relevant literature pertaining to suicidal CYPs views existed which could be synthesised to explore their views of mental health support services.
This study is the first meta-ethnography exploring the views of suicidal CYP in relation to their experiences of mental health support services to address this priority knowledge gap. Lachal, et al. (2015) in their meta-synthesis of the perspectives of young people, parents and professionals regarding suicidal behaviors, found that the violence of the suicidal act leads to incomprehension and an inability to empathise with suicidal young people. However, their review did not specifically address the suicidal young persons' views of mental health services.

Methods
Although many methods of qualitative evidence synthesis exist, meta-ethnography is the most frequently cited approach (Dixon-Woods, Booth, & Sutton, 2007;Hannes & Macaitis, 2012;Ring, Jepson, & Ritchie, 2011). Meta-ethnography consists of seven phases: getting started, deciding what is relevant, reading included studies, determining how studies are related, translating studies into one another, synthesizing translations and expressing the synthesis (Noblit & Hare, 1988). Meta-ethnography was the chosen approach because it goes beyond providing a narrative or thematic review of single research studies to produce a new conceptual model or theoretical insight (Noblit & Hare, 1988). This new insight will add to current knowledge informing the design and delivery of future services for this vulnerable population. Specific review questions (phase 1) were:  How accessible are mental health support services to CYP who have been or are suicidal?
 What are the views and experiences of suicidal CYP of the mental health support services they currently receive or have received?
 What do CYP who are or have been suicidal say that they want from mental health support services in future?
Although devised in the 1980s for use in education, developments in qualitative evidence synthesis, and specifically meta-ethnography, mean this approach is well suited to contemporary health research. A protocol for the review was registered with Prospero (CRD42017071126).
Searching for qualitative literature (phase 2) is generally problematic (Evans, 2002) and triangulation of search methods is advised (Shaw, et al. 2004). Specific challenges associated with defining the search criteria for this review included the lack of an internationally agreed definition of adolescence meaning that papers addressing this population can have an upper age range of 18 to 25 years. Additionally, as definitions of self-harm can include suicidal behavior, the term suicide is not always used in article titles. This review used a broad definition of suicidal to include suicidal behavior (any act of self-harm intended to end ones' life whether fatal or non-fatal) and suicidal ideation (thinking about ending ones' life including the planning of suicidal behavior).
These factors were considered when devising our inclusion and exclusion criteria. For example, the age range was set as broadly as possible to ensure all potential studies were captured, and although self-harm was not used as a search term, papers that used this term were screened for relevance. Broad based, thesaurus and free text approaches were used to systematically search (first author, LG) CINAHL, Medline, PubMed, Child Development and Adolescent Studies, and Psych-Info between June and July 2017 using the search terms shown in Box 1.
The reference lists of potential studies were also screened. An internet search of relevant national and international organisations with interests in this area (CYP and their health and wellbeing) was conducted to identify potential grey literature such as research reports. were combined and reported using PRISMA (Diagram 1) (Moher, et al., 2015). Titles and abstracts of possible studies were initially screened (LG) against the inclusion criteria (Box 2).
Articles were limited to English language publications as there were no funds for translation and papers published before 2000 were excluded as they could be considered unlikely to reflect contemporary practice.

Box 2: Inclusion and exclusion criteria
Inclusion criteria:  Studies in English published after 2000.
 Sample population of children and young people (CYP) aged between 5 and 25 years (inclusive).
 Studies reporting views and experiences of CYP who have used mental health support services for suicidality or self-harm including suicidality.
 Primary reports of studies using qualitative approaches and methods e.g. grounded theory, interviews and focus groups, including mixed methods studies.

Exclusion criteria:
 Studies not in English, published pre-2000.
 Studies where most participants were aged under 5 years or over 25 years.
 Studies reporting views and experiences of CYP who have used mental health support services for reasons other than for suicidality or self-harm including suicidal intent or where data relating to CYP and suicidality were not explicitly reported and could not be independently extracted.
 Studies reporting other views and experiences e.g. CYP attempting suicide, preventing suicide in CYP, parents bereaved by suicide, professionals' views of delivering services or CYP use of other services e.g. GP care.
 Not primary qualitative research reports e.g. qualitative evidence syntheses, reviews, narrative reports or editorials.
The full text of studies meeting the eligibility criteria were screened independently by all authors and disagreements resolved through discussion. It was agreed that four studies (Idenfors, Kullgren, & Renberg, 2015;McAndrew & Warne, 2014;Murray & Wright, 2006;Wadman, et al. 2017) met the criteria for inclusion in the synthesis. (Reasons for exclusions are provided in Supplementary Table 2). Although not a pre-requisite in meta-ethnography, all included studies were quality appraised (Critical Appraisal Skills Programme (CASP), 2017) by first author (LG), and were found to be of either "good" or "high" quality (Supplementary Table 3). This systematic appraisal supported the essential close reading of included texts required in phase 3.  (Noblit & Hare, 1988)) were then extracted verbatim for each study, imported into NVivo, and descriptively coded line-by-line using the review questions as an a priori coding frame (LG). Included studies were related (phase 4) through their characteristics and first and second order data. Translation of included studies (phase 5) was a) reciprocal -through identification of shared themes, concepts and metaphors and b) refutational -by identification of disconfirming cases (Noblit & Hare, 1988). Translation was a complex iterative process involving constant comparison between studies and development of concept maps to create a new line-of-argument. Translation enabled findings to be synthesised into a third order, over-arching reviewer interpretation, going beyond simple narrative reporting of original themes (Noblit & Hare, 1988 (2017) focused exclusively on self-harming CYP who were 'looked after and accommodated' by social work services but used a definition of selfharm to include suicidal intent. All four studies used interviews for data collection.
[Insert Table 1 here] Included papers were reciprocally translated against the review questions as follows:

Challenges accessing mental health support services
Translation of first (Idenfors, et al. 2015;McAndrews & Warne, 2014) and second order data (Idenfors, et al. 2015;McAndrews & Warne, 2014) revealed that CYP found services difficult to access; not knowing where, or how to access help, and being frustrated at not being able to access services directly themselves. For example:

"I did not have a clue in the slightest, I didn't know there was [a] Child and Adolescent
Mental Health Service (CAHMs) (Fiona)" (McAndrews & Warne, 2014:575).
Accessing services was complicated, and often involved young people being signposted to other services or re-referred to services that had not met the needs the first time.
"… I didn't think that they could help me. So, I didn't really trust them. But then I called the mobile team here, a while ago. But then they sent me to the young adult health center. And then I gave it up again (Participant 7)" (Idenfors, et al, 2015:182).
Both Idenfors, et al. (2015) and Murray & Wright (2006) identified that having immediate access to help is an important issue. For example: "Just the fact that I know that I did not come directly to the child and adolescent psychiatry clinic. And that alone is probably difficult, I think. That there isn't a direct number. (Participant 3)" (Idenfors, et al., 2015:181).
CYP participants relied on others, usually friends, family and teachers (Idenfors, et al. 2015; McAndrews & Warne, 2014) to facilitate referrals to mental health services. Wadman, et al.
(2017) noted that the removal of 'looked after and accommodated' children from these generic support networks may impact negatively on their mental health (Wadman, et al, 2017). Given that CYP rely on other adults to make referrals to support services on their behalf, being taken seriously and feeling supported by these adults could be the difference between them going on to access services or not. McAndrews & Warne (2014: 574-575) highlighted that the support CYP received in school was not always appropriate, for example: "…it was brushed off (Nina)" and "I did tell them I was depressed and I had these suicidal thoughts,… but she never said anything (Julie)".
Primary data in Idenfors, et al. (2015); McAndrews & Warne (2014)  Translations based on second order constructs across the studies (Idenfors, et al. 2015; McAndrews & Warne, 2014; Murray & Wright, 2004;Wadman, et al. 2017) reflected the primary data. That is, CYP not wanting or feeling able to talk because they have no trust in services, feelings of shame and perceptions of stigma associated with their suicidal and self-harming behaviours, fear of being judged, and fear of the consequences of disclosure. Idenfors, et al. (2015) refers to attitudinal barriers to help seeking which were identified amongst the young people. Idenfors, et al. (2015) and Wadman, et al. (2107) also identified that many CYP believed that they should help themselves.

Experiences of mental health support services
Reciprocal translation also revealed that when CYP managed to access mental health support services their experiences were variablea "mixed bag" according to Wadman, et al. (2017). Idenfors, et al. (2015) reported that the profession of the person delivering the service is less important to the CYP than being able to get on with them. Positive experiences of care across the studies were ascribed to how the person (professional) made them feel.
""she went out of her way to make me feel comfortable, and I never felt like I was talking to a professional, she'd always make me feel like she was, like she was really, she was so good" (ID 29)" (Wadman, et al. 2017: 9). However, these positive experiences were not, as strongly reflected in the data across all the papers. Two studies (Idenfors, et al. 2017;Wadman, 2017) reported that CYP talked about not being listened to, being patronized, and that therapists/workers failed to engage with them regarding their worries. Therapists/workers often imposed their expert opinion upon what they believed CYPs problems to be, rather than listening to what was being said. For example: "…the lady I was talking to was, she was nice, but she was incredibly patronizing." (Wadman, et al. 2017: 8) "…she doesn't listen to what I say…I don't know she twists things I say to … I don't know how to explain it it's like nothing I say is important" (Wadman, et al. 2017: 8).
"Negative experiences included medication prescribed for depression or anxiety that did not work well, or a regular contact that did not help" (Idenfors, et al.-, 2015:182).
Original author interpretations of these negative experiences did not go further than describing what they had heard from participants and are therefore generally reflective of the first order constructs reported above. But, together they suggest a bleak picture of how this vulnerable population of CYP feels about how they are dealt with by mental health support services.

What CYP say they want from support service
The CYP who participated in these studies had a strong sense of what they wanted from a service; primarily someone to listen to them and treat them with respect. Participants in all studies reported that successful engagement with services was attributable to the individual workers that they met and how they were treated by them.
"I had very high standards for people…if they looked down on me or said something that made me mad, just something a bit off beat, it would completely ruin it. I'd want to leave.
I never wanted to go see somebody and I never wanted to make it feel like they were just there to do their job, and they were getting paid for what they were doing, and they never gave a shit outside of it," (Murray & Wright, 2006: 161) "But then I began to understand that maybe the problem isn't that the help doesn't work, but it's that you have to meet the right person, quite simply (Participant 3)" (Idenfors, et al, 2015: 182).
Although CYP wanted different ways to directly access services, such as by telephone or in person (Idenfors, et al, 2015), most wanted face-to-face contact (McAndrews & Warne, 2014) with someone who listens and takes them seriously (Idenfors, et al 2015). There was only one reference to a young person who wanted on-line support (McAndrews & Warne 2014) and one other expressing a preference to initial email contact (Idenfors, et al. 2015). (2006)  The four studies in this synthesis were conducted in three countries and all were contextually different in their settings (Table 1), and yet, reciprocal translations of the experiences of these CYP were markedly similar in terms of the concepts, themes and metaphors. context. This was the smallest study (n=3) and those with negative experiences may not have come forward for interview. Nevertheless, the descriptions of the specialist nurse actively listening and taking these CYP seriously also supports what participants in the other papers reported as wanting from services (Idenfors, et al. 2015;McAndrews and Warne, 2014;Wadman, et al. 2017).

Line-of-argument synthesis
Through deep immersion within the narrative of these studies and their translations an unexpected finding emerged -there was a distinct lack of discourse about 'suicide' within their content. This was confirmed through linguistic analysis (Biber, Conrad, & Reppen, 1998) Table 5). Linguistic analysis revealed that references to suicide/suicidal in these papers were mostly used in a descriptive context such as referring to study populations, defining self-harm or treatment processes. Overall, there was a lack of reference to suicide/suicidal by the CYP within these studies.
Our new line-of-argument synthesis (Figure 1) is that metaphorically, this is potentially what happens with the suicidal concerns of CYPthey remain hidden and un-addressed. Even though self-harm is a strong precursor for suicide, there was a silence around suicide within the conversations the CYP participants reported having with practitioners, and this has also been reflected within these reporting of this research. The relative absence of suicide terminology in the clinical conversations reported, and limited reference to suicide/suicidal in the findings and discussions of these papers, reflects the current dominance of the self-harm discourse. The use of self-harm as an umbrella term to refer to all self-harming behavior, including suicidal intent, can result in suicide being 'lost' in clinical conversations and research narratives. Self-harm as a prolific term which can encompass suicidal intention, intensifies the focus of risk around selfharm but also helps create and maintain a silence around suicide. If this general lack of direct reference to suicide (as found in these research reports) is followed through into professional practice, then it may be that this silence is contributing to CYP not feeling listened to or believed when they do disclose this information to mental health support services.
A possible impact of this silence could be increased risk of suicide, which resonates with the findings in some of the included studies. For example, when McAndrews and Warne (2014: p576) reported these CYP as: "not being able to talk about problems [which] might in turn exacerbate feeling alone, a situation having the potential to lead to suicidal ideation". Similarly, in the case of a young man described by Idenfors, et al. (2015: p182) who was reported to have taken an overdose after being re-referred to a service that he did not believe helped him first time around.

Discussion
It is known that CYP experiencing mental health problems have difficulties accessing services (Gulliver, Griffiths & Christensen, 2010), and that most CYP who self-harm do not access support (Chandler, 2017  complex 'revolving door' process experienced by CYP accessing services (including poor signposting and the need for re-referral) in which they had to rely on others to access these services for them. Not only do CYP experience barriers in initially accessing mental health support services, once they get referred they can experience further barriers within these systems, such as not feeling listened to or understood by practitioners.
Some positive aspects of mental health support services for these CYP were reported by participants such as; the use of a new risk assessment model that emphasised young people's social context (Murray & Wright, 2006) and the benefits to CYP of meeting staff who they felt empathised with them. However, in general, the views and experiences of CYP presenting with suicidality indicated that the services they received were not adequately meeting their needs and there was an urgent requirement to improve these services.
It should also be noted that despite recent calls to further develop the potential of on-line resources, there was some data to suggest that children who have been, or are suicidal prefer face to face support (McAndrews & Warne, 2014). The importance of connecting with real people is reflected in the wider suicidolgy literature (Lakeman, 2010).
The need to hear the voices of children with complex mental health issues has been recognised generally (Hart & O'Reilly, 2017) and specifically in the context of suicide (Bergmans, et al. 2009;NSPCC, 2014). There is a need for healthcare practitioners to listen to the views of suicidal children when delivering care (Montreuil, Butler, Stachura, & Pugnaire Gros, 2015).
However, our line-of-argument synthesis is novel, enabling a more nuanced understanding of such issues by suggesting there is a potential to silence the subject of suicide in clinical conversations when young people do not feel listened to. Our meta-ethnography has also revealed further original insight by highlighting other factors that undermine or unintentionally silence the issue of suicide in CYP for example, the lack of exploration or references to suicide by academic authors conducting and reporting research in this field (Supplementary Table 5).
Despite suicide in CYP being an international concern, it was also difficult to find studies reporting their views and experiences, for inclusion in this review. The use of the umbrella term 'self-harm' to include suicidal behavior within the literature contributed to this but it was also because some studies which included child and adult participants did not report data separately for these two groups, resulting in the voices of suicidal young people being lost within the overall reporting of data.
The voices of specifically vulnerable and at-risk groups were also notably absent from the Finally, the findings of our meta-ethnography suggests that silence around suicide may also extend to conversations between CYP and their parents or in parental attitudes as illustrated by parents declining consent to their children taking part in research saying, 'my child is doing very well now and I do not want to bring back their suicidal thoughts and behaviors by talking about their experience' (Murray & Wright, 2006: 159). The UK charity ChildLine has also reported that for many of the children contacting its helpline, their parents had declined to have any further discussion with them on the topic or tried to change the subject (NSPCC, 2014: 24).
Overall, this potential parental, practitioner and academic silence around suicide in children is an issue that warrants further investigation. It may be that there is some difficulty in comprehending or acknowledging suicide in CYP and self-harm is somehow easier to deal with.
It may be that there are fears that talking about suicide may contribute to more suicidal thoughts and behaviours by CYP. The reasons behind parental silence may be different from those of professionals or academics, and therefore more research is needed to understand these phenomena from different perspectives.

Practice implications
Mental health support services for use by CYP with suicidal behaviour and ideation need to become more easily accessible for this population. Consideration should be given to providing services that CYP themselves can directly access and which encourage access by marginalized groups; remaining mindful that there is some evidence to suggest that children and young people prefer face to face support.
Practitioners need to be aware of, and address, barriers during their consultations such as CYP not feeling respected or listened to and be aware of the potential to silence suicide in clinical conversations (and research), especially when they may miss dealing with the risk of suicide by exclusively adopting the language of self-harm.

Limitations
Identifying literature that solely addressed suicidality in CYP is made more difficult because selfharm definitions can incorporate suicidal behavior. Our electronic search focused specifically upon suicide and this may have limited the number of studies that we were able to include because we did not purposively search for studies that used self-harm terminology to include suicidal behaviors. There is a significant volume of self-harm literature. To identify the views of CYP who were suicidal within those who studies self-harm within studies that combined suicide within the term 'self-harm' would have required substantial resources that were beyond the scope of this unfunded study. However, our study highlights the scarcity of literature specifically addressing the views of suicidal CYP in relation to mental health services. A further limitation to our search criteria was that we were only able to include those studies published in English and which were accessible in our timescales.
The included studies did not fully describe the gender identity, ethnicity or socio-economic status of their participantsthe ethnicity of participants was only stated in one study. The participants in all of these studies were disproportionately female and living in Europe, therefore our findings may have limited transferability to other more vulnerable groups.
Studies included in the meta-ethnography were conducted in three different countries.
However, despite international differences in service provision, their findings shared many similarities such as these CYP needing to know where to access help themselves and the importance of needing staff who understood them and they can connect with (Table 1).
The purpose of a meta-ethnography is to elicit new theories / understandings and this can be achieved with a small number of studies, what is critical is the conceptual richness of the data as this is essential for translation and synthesis (phases 5-6) (Noblit & Hare, 1988). Data extraction confirmed that all four studies reported data of the necessary depth required for translation and synthesis. However, our line of argument, as with any new theory arising in this way requires testing through further research.

Conclusion
This is the first meta-ethnography of studies reporting the views and experiences of CYP who have used mental health support services for suicidal behaviors. As such it breaks new ground and provides novel insight into mental health support services from the perspective of suicidal young people who have been, or are, using these services. There are both internal and external barriers to CYP accessing mental health services for suicidal behaviour. Children and young people who are suicidal need to be aware of and have easy access to mental health services.
When using these services, they want to feel listened to and have their suicidal feelings acknowledged. This involves professionals referring explicitly to suicide and not just self-harm.
There is a potential silence around suicide in conversations between CYP and mental health practitioners. Research in the field may also perpetuate this silence, especially when researchers do not differentiate between suicidal and self-harming populations.
Future research is needed to investigate and ascertain these translated findings and new lineof-argument from the perspective of a larger number of CYP participants especially those from marginalized groups at greater risk of suicide. There is also a need to explore our findings from the perspectives of the practitioners involved in delivering mental health services, and potentially researchers in the field.

Acknowledgements
LG is an ESRC funded PhD student. No other funding was received for this study. All authors contributed to the study design and conduct.
LG led on all stages with NR and MM crossvalidating and contributing to the translation and synthesis process. The authors declare they have no competing or potential conflicts of interest. All authors contributed to writing of the paper.

Supporting Information
Supplementary

Phenomenological analysis
In-depth interviews in interviewer's office.
Interviews conducted by 2 nd author who had no relationship with participants Parents were also involved in interviews as this was part of a larger study, relating to a new model of suicide risk assessment, with parents' perspectives reported elsewhere.
Themes from 1 st order participant data: YP spoke of fear, anxiety, and shame after being referred for suicide risk assessment; the importance of a quick referral process, immediate help and having one therapist over a consistent period to establish trust in their relationship with the practitioner.
Change -Three different aspects were reported after the assessment process: change in their thinking, other intrapersonal changes & a change in interpersonal relationships.
Hope -that things could or would change and how they shifted from a sense of giving up hope to hopefulness. Recognition of their own ability to deal with their problems/issues and the importance of not feeling alone.
Connection -the importance of the connection to the clinical nurse specialist and the importance of feeling that someone cared for them. Study aim: to present narratives of YP with self-harming and or suicidal behavior, to identify how services can best meet their needs.

1-1 interviews at venues to suit participants.
Themes from 1 st order participant data: 1. Cutting out the stress e.g. triggers of self-harm 2.
Stepping onto the path of help e.g. deciding to access help 3. Cutting to the chase: prioritising selfharm on the public agenda e.g. knowing who can help.
Themes form 2nd order author interpretation: -Many things trigger self-harm, and YP often use self-harm as a way of coping -YP experience feelings of shame associated with their SH behaviour.
-YP find it difficult to identify who they should talk to about their self-harming or suicidal behaviour.
-Young People identify that they prefer face to face contact, with someone independent from family and friends, who displays a nonjudgmental attitude, listens and is someone they can trust.
-Schools are well placed to support CYP around the onset of self-harm. Themes f orm 2nd order author interpretation: -There is a need for more flexible, available, and varied health care.
-Suicidal CYP struggle between wanting to be independent and needing reliable support.
-Contact with CYP that are suicidal should be empathic and demonstrable active listening. -Involvement of family and friends is crucial and might be pivotal in supporting access to services. Wadman, R, Armstrong, M, Clarke, D, Harroe, C, Majumder, P, Sayal, K, Vostanis, P, Townsend, E 2017 n=24. 14-21 years Male =4; Female =20. Ethnicity: not described.
All were looked after CYP or recent care leavers who had self-

UK -East Midlands
Participants had all been looked after and accommodated.
Study team: multidisciplinary; included professionals from Interpretative phenomenological analysis.
Semi-structured interviews in various locations. Interviewer from a research not clinical background. Emotional rating scales completed before and after Themes from 1 st order participant data: -Changes in placement are perceived as highly relevant to SH either as a cause or consequence. -Feelings of anger (and turning anger on self). -Not wanting to talk; not feeling able to talk. -developing their own coping techniques to deal with SH.
-experience of clinical services: a relational mixed bag.