“I’m still here, but no one hears you”: a qualitative study of young women’s experiences of persistent distress post family-based treatment for adolescent anorexia nervosa

Background Family-based treatment (FBT) is the current treatment of choice for adolescent AN based on positive outcomes that include weight restoration in around two-thirds of adolescents. Nevertheless around a quarter drop-out from treatment, particularly in the earlier phases, and a notable proportion of treated adolescents are reported to experience ongoing psychological distress during and post-treatment. This study explores the under-researched experiences of these adolescents. Method Fourteen participants from Australia, New Zealand and the United Kingdom were interviewed about their experiences of FBT. An inductive thematic analysis of interview transcript data generated key themes related to their experiences, identity negotiations and the discursive materials these used to construct these. Results The participants identified working as a family unit as key to their recovery, highlighting the importance of family therapy interventions for adolescent AN. However, they perceived an almost exclusive focus on weight restoration in the first phase of FBT was associated with experiences that included a relative neglect of their psychological distress and a loss of voice. Key within these experiences were processes whereby the adolescent engaged in identity negotiation and (re)claiming of their voice and implicit in their family standing with them in the treatment was that their life was worth saving. What was noted as most helpful was when therapists advocated and took into consideration their unique needs and preferences and tailored treatment interventions to these. Conclusions There is a need to develop and research treatments that address, from the outset of treatment, the adolescents’ psychological distress (including as experienced in the context of their weight restoration). This should be with priority accorded to the adolescent’s voice and identity negotiations, as they and their families take steps to address the physical crisis of AN and in doing so, support more holistic and durable recovery.


Background
Anorexia Nervosa (AN) is characterised by significantly low weight and an obsessional fear of gaining weight [1], frequent onset in the adolescent years [2] and mortality of 6-15%, with half of all deaths resulting from suicide [3][4][5][6]. A complexity for the treatment of AN is the egosyntonicity of the AN symptoms, many of which are valued by the experiencing person that contribute to a reluctance to engage in treatment [7]. For those who do engage in treatment, drop-out rates are high, including 23-73% in outpatient settings [8].
Family Based treatment (FBT) [9] has been reported to have positive treatment outcomes, particularly on eating disorder (ED) symptomatology, including in those adolescents assessed with higher ED symptomatology [10]. FBT has been proposed as the first line, 'gold-standard' evidence-based treatment for adolescent AN [11]. Through a number of randomized controlled trials (RCTs), FBT has been found to be associated with earlier weight restoration and reduced ED symptomatology and lower hospitalizations (and hence treatment cost), compared with adolescent focused individual therapy [12] and family therapy that addressed systemic concerns rather than eating related behaviours [13]. Further RCTs have found parent focused treatment (PFT) to be more efficacious than standard FBT; PFT being where treatment consists of the FBT therapist meeting with the parents separately to the adolescent who meets with a specialist nurse [14]. There is also evidence to indicate that when parental expressed emotion is higher, PFT results in higher rates of remission in the adolescent [15]. Early weight gain, a specific focus of FBT, has also been found to predict improved outcomes for the adolescent, particularly when there is a corresponding early focus on addressing negative emotion, parental criticism and the therapeutic alliance early in FBT and PFT treatments [16]. Furthermore, therapeutic alliance, as measured by an independent researcher-rater using the Working Alliance Inventory Observer Version (WAI-o) [17], has been found to be 'achievable' [18,19] and 'generally strong' [20] in FBT. However, this measure of therapeutic alliance in these studies did not find a significant relationship and treatment outcome of remission of AN symptoms [18].
FBT is a three phase manualized treatment that begins with a focus on supporting parents to take responsibility for their child's eating and weight restoration, progressing to phase 2 where this responsibility is re-allocated to the adolescent once they have increased their food intake and sustained reliable weight gain. The final phase commences when the adolescent maintains at least 95% of their ideal body weight, and is focused on re-establishing autonomy and developing a healthy identity [21,22]; psychological processes and relational problems are not specifically addressed until this final phase of treatment [23,24]. These phases of treatment are estimated to be delivered over 12 months over 20 sessions. Recent evidence in clinical settings has demonstrated that for those who do not meet remission by session 20, that further FBT sessions resulted in lower number of days in hospital and improved rates of remission compared to those who sought out alternative treatments to FBT [25].
A Cochrane review of 25 published and unpublished randomised controlled trials that compared family therapy interventions, including FBT, with other AN treatments [26] has found some evidence of a small effect size of FBT related to weight gain post-treatment, however, little evidence of differences between outcomes for groups across all comparisons of treatments, including in ED symptomatology (weight, ED psychopathology), drop out, relapse or family functioning measures at post-intervention or at follow-up. This review concluded that there currently exists insufficient Plain English summary Family-Based Treatment (FBT) is a well-established, intensive approach to the treatment ofadolescent anorexia nervosa (AN). The first phase of treatment focuses on eating and weightrestoration, where parents are given responsibility for the adolescent's home-based refeeding.This is followed by handing over of this responsibility to the adolescent with the final-phase afocusing on adolescent-specific psychological issues. While the majority of adolescents gainweight with this treatment, a substantive proportion experience ongoing psychological distressand around a quarter of families drop out in the first phase. Little is known about theseadolescents' experiences.
In this project, we interviewed 14 individuals who, with their family, had either dropped outof FBT and/or experienced ongoing psychological distress post-treatment. The participants notedthat their parents' stance in supporting them was life-saving and contributed to a sense that theirlife was worth saving. However, they also noted a relative absence of focus on theirpsychological distress, particularly in the early stages of treatment. Most helpful for participantswas when therapists took into consideration and tailored treatments to them. Future treatmentsneed to consider ways to support an adolescent's psychological distress more comprehensively,prioritise their voice and support them in finding an identity outside of the AN identity.
evidence that family therapy interventions were superior to educational interventions and other types of psychological interventions for adolescent anorexia nervosa at long term follow up. There was also insufficient evidence of (1) outcomes outside symptom remission with only two trials of 21 reporting outcomes based on a return to normal functioning; and (2) whether one type of family therapy was more effective than others. Furthermore, the lack of detailed information about individuals and their families who drop out of adolescent AN treatments may have contributed to "the effect of artificially inflating the effectiveness of the interventions reviewed" [26].
Overall, results from RCTs show that FBT works well for less than half of the adolescents and their families who engage in this treatment [27]. Outside of this group for whom this intervention works well: (1) up to 27 percent of families drop out from FBT [28]; (2) 40% of adolescents continue with ongoing psychological distress despite weight restoration [29]; and (3) comorbid symptomology has a negative impact on treatment outcomes and increases rates of attrition [30][31][32]. In response to these findings, and given some promising findings for enhanced cognitive behavior therapy (CBT-E) adapted for adolescent AN [33,34], CBT-E has recently been compared with FBT in a non-randomised effectiveness trial [35]. Findings indicated that greater early weight gain in FBT that by 6 to 12 months was indistinguishable between the two groups. Both treatments resulted in similar weight gains in general measures of psychopathology and clinical impairment; keeping in mind that the group who elected CBT-E were older and more unwell. Those who fared less well with these treatments reported a history of abuse and, specifically in the lower weight cohort, reported higher internalization, a presence of comorbid psychopathology, and prior mental health treatment.
A meta-synthesis of qualitative research into the experiences of family therapy for AN [36] has found three of 15 papers analysed adolescent and family experiences of FBT with one consisting of data generated from openended survey questions [37]. Two of these studies [37,38] found that the most helpful aspect of FBT was parental support and understanding and the third study [39] found that the authoritative stance of therapists having an oversight of their eating behavior to be helpful. Less helpful was the neglect of issues other than AN [37,39] and the adolescents' unmet preferences for individual therapy [37]. There is increasing evidence to support the need for ED treatment interventions to focus on more rapid relief of the psychological dimensions of AN [40]. Furthermore, research into treatment experiences of those with a lived AN experience has identified a need for treatments to more directly address questions of identity, including the rebuilding a sense of identity outside of the ED identity [41,42].
Overall there have been few studies that have focused on the experiences and identity struggles of those whose families either drop out of FBT and/or adolescents who continue with substantive psychological distress posttreatment [29]. This study sought to address this gap and give voice to those who experience persistent distress post FBT.

The current study
The current study utilised a qualitative framework to give voice to the person with a lived adolescent AN experience with a focus on participant: 1) Experiences of Family-Based Treatment for adolescent AN in those who were interested in participating in research to improve the intervention; and 2) Identity negotiations in the context of this treatment intervention.
The aim of this study is to explore how these participant experiences and identity negotiations might inform future augmentations and transformative treatments for adolescent AN.

Design
This study was an inductive thematic analysis [43] with the understanding of themes as constructed within an interpersonal context. Analysis of data that comprised the themes/subthemes focused on some of the ways participants negotiated their identities through the discursive materials or language forms available to them at the time, the semantic and latent meanings they ascribed to the experiences of FBT and some of the dilemmas they faced [44] in the context of their recollection of experiences of FBT for adolescent AN. This study was also part of a larger study that explored the experiences and perspectives of parents who had experienced FBT [45] and clinicians who reported being trained and having practiced as FBT practitioners [46].

Participants
A purposive sampling technique was utilised to invite participants to talk about their experiences of FBT and generate a context through which they could voice aspects of the treatment that was both helpful and their ideas of ways the intervention could be improved. These participants responded to advertisements via Australian clinicians; and after indicating their interest on completion of an eating disorders (ED) treatment experiences survey advertised through Facebook (see "Appendix A" for the wording of the research advertisement).
Fourteen participants aged 14-27 years (M = 18.58, SD = 3.20), who reported being diagnosed with AN and treated with FBT for adolescent AN on average 4 years earlier (range 1-14 years) were interviewed (in person or online/telephone). See Table 1 for further demographics and FBT treatment details. Half of the participants reported weight restoration post-treatment (with one later relapsing); all the participants reporting ongoing ED symptoms and psychological distress post-FBT. Eleven participants reported additional treatments for co-morbid psychological problems prior to or post FBT with three of these participants reporting psychological counselling whilst also engaging in FBT (see Table 2 for details).

Procedure and materials
This study was approved by the Western Sydney University Human Research Ethics Committee (approval number: H11303).
Semi-structured interviews were carried out by two researchers (JC, SC) (see "Appendix B" for interview format) that were audio recorded, transcribed verbatim and de-identified with participants' chosen pseudonyms. Transcripts were then given to participants to member check for accuracy and the removal of further identifying information for confidentiality.

Analysis
The analysis was data-driven and inductive with all themes generated from the dataset [43]. Throughout the analytic process, themes were constructed through the explicit language used by participants, and an analysis of implicit meanings in their narratives, as language in this instance is assumed to construct a version of the participants' experiences and meaning-making processes [47]. Three authors (KS, SN, JC) familiarised themselves with the data by reading and re-reading transcripts, coding meaningful units of raw data into draft themes and subthemes. Two authors (CJ, JC) also coded raw data into 'nodes' using QSR NVivo-12 qualitative data analysis software, examined the nodes for similarities and differences and grouped data related to the research question into categories. These categories of data were then collated and contrasted with the earlier drafted themes/ sub-themes before being analysed for any further relationships and grouped together to generate overarching themes and a thematic map that addressed the research questions through in-depth analysis of exemplar data within each theme/sub-theme (JC, CJ, PH). Analysis by the researchers (see "Appendix C" for researcher positioning statements) traced some of the discursive materials these participants used to piece together key experiences and dilemmas associated with components of FBT and including patterns of identity negotiation [44,48,49]. The draft analysis was given to participants to member check for the purposes of validity (see "Appendix D" for member feedback provided by 4 participants) and to align analysis with participant feedback.

Results
Analysis traced these participants' experiences of key dimensions of the FBT intervention and ways they engaged in identity negotiations within these treatment contexts, including the reclaiming of identity and voice in matters related to their treatment (see thematic map, Fig. 1).

Theme 1: Therapeutic focus
Participants recounted being both supported and distressed by the focus of Phase 1 of treatment on eating and weight restoration and handing control of these over to their parents. Of the 4 participants who reported completing the three phases of FBT, 3 concurrently had treatment with an individual therapist (Abbey, Amy & Maisy) and one participant retrospectively described being "taught" strategies by psychologists and dietitians during treatment to assist her with managing distress (Phoenix). Participant narratives highlighted that although FBT was preferred to inpatient treatment for most, when the therapeutic focus was predominantly on AN symptom reduction their emotional distress was obscured and further escalated.

Subtheme 1(a): Focus on the visible
Ten participants' narratives indicated that the early treatment focus of FBT, where their parents took responsibility for their eating, was experienced as a relief, albeit it was also a distressing experience.

EXTRACTS 1
Pheonix: […] As much as I hated it, and as much as I just wished for it to be over, it (FBT) did save The allocation of responsibility for their eating to their parents was experienced as life-saving for these participants. The marking out of a boundary by which they were no longer responsible for their eating cultivated for some, a sense of relief and safety (e.g. Kaylee) and the possibility of getting "your life back" from AN, despite also being "extremely distressing" (Phoenix). Furthermore, the sense that others could "be my eyes" (Kate) indicated that these participants valued not only the support but also the alternative perspectives of others in assisting them to take steps to diminish the influence of AN on their lives. This included parents being supported by treatment teams to engage in nutritional nourishment despite a divided sense of self ("part of me … fighting it"; Abbey). Identified as helpful was when their parents were experienced as a steady support, informed yet understanding, cultivating safety whilst taking control over their eating that contributed to a containment of their distress in the context of the challenging time of nutritional restoration.
Alongside FBT being a life-saving treatment that addressed the medical crisis of AN, the participants also recounted parallel experiences of distress during and post FBT irrespective of extent of their eating and weight restoration and all of them acknowledged the significance of support by others for this distress, including family members.

Theme 1(b): Focus on the invisible
Thirteen of the 14 participants argued that there was a lack of treatment focus on their experiences of psychological distress during FBT, including co-morbid psychological problems (see Table 2). The remaining participant (Maisy) reported FBT was experienced as "extremely pivotal to me getting better" and tailored to her preferences where her siblings attended only the first session. Individual therapy in addition to FBT was experienced for her as both an opportunity to "challenge my thoughts", "learn how to self-soothe" through mindfulness techniques and a motivational intervention that worked "to convince me that recovery was a good thing". She also highlighted the significance of seeing an individual therapist in addition to FBT who was "an outside person […] different from my parents". For ten of the participants who reported dropping out prematurely from FBT, an absence of focus on addressing emotional distress and interpersonal struggles was cited as a major contributor to this decision. The inadvertent effect of prioritizing of physical safety in phase 1 of FBT meant that these participants' psychological safety was obscured with the implicit meaning taken from what was "going on inside" (Hayley) did not matter. For Charlotte, the structure of FBT meant the "old wounds" were opened and not healed, including the parallel process of eating difficulties of women in her family being left unaddressed contributing to an identity conflict where she was left questioning herself and to feelings of "sadness". Implicit in these participants' experiences of anger, was their valued stance and desire for openness within their family systems. They recollected their anger being responded to by being positioned as outsiders to their own therapy, losing their voice with systemic family issues left substantively unaddressed.
Within this context, a different sort of safety issue arose with six of the 14 adolescents reporting suicidal ideation during and/or post FBT treatment. These participants argued that nutritional and weight restoration alone did not reduce their psychological distress; distress that continued largely unaddressed for them during the FBT intervention. Kate recollected experiences of a loss of voice where she was recruited into need(ing) "to sit there and be quiet"; Beth actively hid her distress, particularly from her mother; and Harley's feedback on the analysis of her transcript (member checking) emphasized the lasting impression of treatment where being talked "at" or "about" continued to trigger "suicidal ideations" with continued identity associations of herself as a "burden" on others. These participants were active in their arguments that AN treatments need a greater focus on enabling the voicing, holding and processing of their and their family members' emotional distress.
Parallel to the adolescent's distress was family distress and conflict that was also experienced in the context of parents taking responsibility for their eating and weight restoration.  Increased conflict in the family was evident in the participant narratives, with examples cited including angry outbursts with four participants disclosing anger that was physically expressed through parental restraint such as being held down or against a wall. Kaylee clarified that she was not blaming her parents for their responses at the time, that included physical restraint, however, she also recounted how these responses contributed to further distress, including suicidal ideation.
On the other hand, some of the participants talked about their experiences of FBT therapists advocating for them and mobilizing their parental support in ameliorating a sense of being "alone" in addressing AN.  These extracts exemplify key components of the therapeutic alliance that included FBT therapist advocacy, scaffolding insight into themselves and others' understanding of their experiences (Amy) and ameliorating a sense of being alone in treatment through engagement with families in the intervention (Beth), including in standing together against AN (Charlotte).

Theme 2: Identity negotiations
Alongside the participants' treatment experiences were their parallel identity negotiations that were shaped by their experiences of personal agency and voice throughout treatment and their engagement in the FBT practice of externalisation of the illness (theme 2a). Furthermore, the implicit meanings ascribed to their parents' support was that their life was worth living (theme 2b).

Theme 2(a): Negotiating personal agency and voice
All the participants, at some point in their narratives, talked about struggles to negotiate personal agency and voice in their treatment, particularly in the first phase of FBT where their parents were allocated responsibility for their eating restoration.  These extracts exemplify the identity negotiations of participants as they ascribed meaning to their parents being asked to take responsibility for their eating in the first phase of FBT. Phoenix's use of the metaphor of a "monkey in a cage" depicted her experiences of a loss of personal agency, being monitored and unable to escape. The participants ascribed a number of negative identity conclusions in the context of their parents taking responsibility for their eating-for example for Kate this reinforced the identity of herself as a "sick" person that did not fit with who she understood herself to be.
Furthermore, parent's taking responsibility for their eating had real effects on some of the participant's relationships with their parents that had unintended impacts on their sense of themselves as a daughter.  These extracts exemplify the potential effects of the FBT practice of externalisation of the illness that led to the person's voice being assumed by others to be the AN/ED. This misappropriation of the person's sense of self ("I'm still me"; Nora) to the disorder by others ("I was an eating disorder"; Lydia) contributed further to a loss of voice and exacerbation of distress. Charlotte traced the problem of externalization when built on the assumption that it is possible to achieve "separation" of the person from the ED when their identity is invested in the egosyntonic dimensions of the experience ("I felt that was my identity"). These extracts highlight the unintended consequences and struggles when the practice of externalization of the illness aimed to completely separate AN from the person and neglected to take into consideration the identity investments into the egosyntonicity of AN.
On the other hand, five of the participants found the process by which their therapists engaged them to externalize AN/ED to connect with a sense of identity outside the AN.

EXTRACTS 9
Phoenix: […] as I was restoring weight and as I was getting better and given more privileges and so on that I got to really find out who I was.
[…] they made us draw a Venn diagram with two circles. And they named one side "Phoenix" and one side anorexia and then throughout treatment they would make me draw where I thought the circles were overlapping and there was definitely a correlation between the distance of those circles and the amount of weight I restored. As  Emerging across these participant narratives was their preference for a person-centred approach where the practice of externalization was focused on the person's identity outside of the AN rather than primarily on elimination of the AN. Phoenix recounted a process of "finding out who I was" through the process of being given "privileges" as she gained weight and tracing her shifting relationship with AN as both individual and overlapping entities. Rachel was invited through a chair technique to have a dialogue with AN to enable her to reclaim her voice and preferences from AN; she argued that her preference was to do more of this work that indicated care for her feelings than being centred on eating. Kate outlined how therapy sought to enable her to distinguish between the voice of the ED and her own voice and Abbey talked about how externalization enabled her to reclaim both her voice and identity outside of the AN identity.

Theme 2(b): Life is worth saving-"No one was ever going to give up on me"
Interwoven in participant narratives was a process of ascribing meaning to their parent's commitment to their recovery. The participants ascribed a range of meanings to their parents' support during their treatment, including taking responsibility for their eating in the early phases of FBT. Recollections ranged from parental support and reassurance (Phoenix) and parents' capacity to be with them when emotionally distressed (Kate). Reflecting back, Jessica connected with her parents' efforts to "understand" her experience of AN and Maisy with her parents' perseverance to "get me where I am today". Furthermore, three participants (Kate, Maisy & Abbey) specifically remembered their parents taking up this role in treatment to avoid them requiring inpatient treatment. These participant experiences highlight the importance of parental support in AN treatments.

EXTRACTS
For participants who discontinued FBT in the earlier phases, parental support was also noted in their parents' active collaboration with them to find alternative AN treatments that met their needs and preferences, including treatments that focused on addressing their psychological distress.

EXTRACTS 11
Lydia: I had some serious conversations with my parents and I think that I began to sort of get through to them and I think they to some extent also realised that this really wasn't working, and I needed a different sort of treatment, a different sort of support than what I was receiving. Nora: […] I would say that our relationship [with her mother] now is better than ever and we are able to reflect on the experience and how traumatic it was and how much we both believe the more sick I am is quite harmful to our family and how great it was when we did get individual psychologists.
Implicit in both these extracts was the significance of these participants being validated in their treatment needs and preferences and specifically for Nora in "how traumatic" AN and its treatment with FBT and hospitalisations were for her and her family. Furthermore, all the participants talked about the significance of their parents standing for them as a person in the face of AN and its treatment.  These narratives highlight the significance of the parents standing for these individuals in cultivating a sense of teamwork (Abbey), care (Amy, Phoenix), and that they were not alone (Beth, Jessica). Implicit in all the adolescent narratives was that their parents' commitment to them and their treatment was hope for their futures and the sense that their lives were worth saving-for example, as depicted by Maisy-"no one was going to give up on me".

Discussion
Family-Based Treatment continues to be the frontline treatment for adolescent AN in Australia, although the evidence for it's effectiveness in addressing both the physical and psychological symptoms of AN is incomplete. This study sought to understand and give voice to the experiences of adolescents who had either dropped out of FBT or continued to be distressed post-treatment to inform future treatment interventions and research. These young women reported that family support in the context of the treatment was instrumental in saving their lives and contributed to the sense that they were not alone and mattered as a person. Nutritional restoration is non-negotiable [50] in the early stages of any AN intervention to prevent potentially adverse medical outcomes [51]; however, this study has highlighted that delaying interventions to address the individuals' psychological distress, including in the early stages of treatment, contributed to a loss of voice and/or an exacerbation of their distress. This is consistent with more recent FBT research that has demonstrated the imperative of early treatment focus on not only adolescent weight gain, but also the therapeutic addressing of difficult emotions, criticism of parents towards their child and the alliance between the adolescent/family and the FBT therapist [16].
The current study found the majority of the participants experienced a loss of voice, particularly in the early stages of FBT. This was evident in contexts where their voice was assumed to be the voice of the illness/AN (by parents and/or therapists) thereby externalizing their identities with the AN [24]. Notable in this process, was a disordering of their identities where they were assumed to be incapable of having a valid voice or perspective on their own lives. Furthermore, the neglect of any substantive focus on their psychological distress in the early stages of treatment set the precedent that their internal struggles were of lesser significance. This lack of focus on their psychological distress was cited as a major reason for treatment drop out. Other responses included anger and disengagement from treatment with the sense of themselves as outsiders to their own treatment. This concern has been echoed by Greg Dring [23]: […] if the therapist spends the first sixteen sessions of the work discouraging the discussion of feelings, relationship issues and developmental difficulties in a personal way, then it may be very difficult to revive such discussion at a later stage when, in any case, the work is about to be concluded (p. 66).
The participants' narratives also exemplified how the early treatment focus of FBT on adolescent nutritional restitution inadvertently risked obscuring not only the young persons' distress but also their family's systemic distress, which had a recursive effect on their emotional distress. The efforts of parents (as supported by therapists) to encourage their child's nutritional restoration therefore, at times, unintentionally may have diverted their focus away from holding their child's less visible, emotional distress. Furthermore, the participants recounted distress and interpersonal strain on their family relationships in the context of parents taking responsibility for their nutritional restoration. It is important to note that some of the parental responses reported by the participants are contraindicated and proscribed in FBT (for example, parental physical restraint). Some of the participants' experiences highlight possible challenges of parents taking responsibility for their adolescent child's eating, including whether the ends of adolescent weight gain justify the therapeutic means by which this may be achieved [24], and the imperative of FBT therapists prioritizing the building of a therapeutic alliance [16]. Important in this alliance is a safe context for adolescents and their families to disclose interpersonal difficulties, and that therapists have skills in addressing family conflict, should this arise. Furthermore, in prioritizing the therapeutic alliance, the implementation of the intervention may align more closely with the adolescents preferences for a greater therapeutic focus on addressing intra and inter-personal distress throughout treatment.
Contrary to expectations, the therapeutic alliance, measured by the researcher-rating of the therapeutic alliance (WAI-o), has not been found to be related to AN symptom remission with FBT in previous research [18,19]. This is a counterintuitive finding when considered with other areas of psychological research where the therapeutic alliance has been shown to predict psychotherapy outcomes more broadly in children and adolescents [52]. The therapeutic alliance is a complex construct and the WAI-o instrument may have been measuring an aspect of the treatment alliance that is not associated with outcomes for FBT with adolescents experiencing AN. At this stage we do not know. It is an interesting finding that needs further research. However, in keeping with others who have highlighted the importance of listening to a person's preferences out of respect for their 'rights and dignity' and to optimise treatment outcomes [53], this present study indicated that participants valued a strong therapeutic alliance, albeit this may not translate into outcomes that requires additional research. Therefore, the differences between this and these previous FBT studies may be the methods of assessing and reporting therapeutic alliance and perhaps the construct itself.
Furthermore, the present study findings are consistent with Sibeoni et al. [54] who found that for 15 adolescents who had engaged in both outpatient and inpatient treatment for AN, the therapeutic alliance was facilitated by: the adolescent perception of the treatment team as authentic where they felt heard and understood, themselves and their parents playing an active role in their treatment, time to develop trust with the treatment team with the adolescent-parent relationship as 'the central element of the therapeutic alliance' . These findings are similar to the experiences of the adolescents in this study, including the importance of a safe therapeutic context to voice their experiences and concerns and to address these, including family difficulties and conflict, should these arise.
The findings of this study are also consistent with Medway and Rhodes' [36] meta-synthesis into adolescent experiences of family therapy for AN who concluded that psychological interventions for adolescent AN would benefit from scope for interventions that focus more comprehensively on underlying adolescent and family issues. This may go some way to ameliorate the ongoing psychological distress post treatment reported in the current and prior studies [29] and may also prevent the progression on to severe and enduring AN for some [55].
Aspects of the FBT experience were also cited as important in participant recovery journeys. These included parental understanding of their AN experience and standing for them as a person, being treated outside of an inpatient setting, and clarifying a sense of identity outside of the AN identity or "find(ing) out who I was" (Phoenix). These participant experiences highlight the importance of addressing questions of identity in AN treatments [41,42] and how externalization as an intervention may, in some contexts, facilitate this process. On the other hand, in contexts where the adolescent themselves was excluded from the process of discernment and naming of the AN, externalisation was experienced as invalidating rather than having the intended effect of empowering the person to reclaim their identity from AN [56].
A number of adolescents in this current study reported comorbid depression, anxiety, and/or OCD and all the participants reported experiencing ongoing psychological difficulties irrespective of weight restoration. This is in contrast to recent research that has shown a significant reduction in co-morbid major depressive disorder, generalized anxiety disorder and panic disorder; proposed in those for whom these conditions are likely to be secondary to malnutrition [57]. Notably, a third of participants retrospectively reported escalating suicide risk during the treatment intervention. This is of concern, given the findings that half of all deaths in AN result from suicide [3,4]. Consistent with previous research [36] and an Australian Broadcasting Commission medical report in 2017 (https:// www. abc. net. au/ news/ 2017-05-04/ austr alianhealth-system-faili ng-patie nts-with-eating-disor ders/ 84853 00), the participants in this study argued for more holistic approaches to AN treatments and that, in their experiences, there was limited scope for FBT to be tailored to their individual needs and preferences, thereby contributing to the decision for some of the participants to cease treatment prematurely and/or continuance of ongoing psychological distress post-treatment.
The importance of addressing an adolescents' psychological distress has been increasingly recognized over the past decade with a number of FBT treatment augmentations being proposed and researched. These have included multi-family therapy [58,59], parent-to-parent consultations [60], separated-family therapy [61], addition of psychological interventions such as CBT [62] and Dialectical Behavior Therapy (DBT) [63] and therapist guided internet chat rooms [64]. These augmentations have tended to focus on changing the context of treatment, providing additional or novel means of support in addition to treatment as usual, particularly for parents [65], rather than augmentation to the structure and/ or transformation of content of the treatment intervention itself. The structural augmentation to separate parent and adolescent sessions, particularly in phase one of FBT, has found similar weight gain outcomes, irrespective of whether the adolescent is involved in the early phases, with increased retention and remission rates for some adolescents [14]. Furthermore, the majority of augmentations involve phase one only and are reflective of the reluctance to change or 'tamper' with the manualized intervention, despite the call for more 'potent augmentations' to improve outcomes [60]. There continues, however, to be a paucity of researched interventions that have scope to transform the landscape of treatment options for anorexia nervosa.

Clinical questions
This research study has explored the treatment preferences of 14 adolescents who either dropped out or continued with psychological distress post FBT and the following questions need to be interpreted in light of this. These participants' experiences prompt consideration of the following questions by clinicians and researchers in relation to future treatment interventions for adolescent AN: (1) What is going on for the adolescent, including their emotional life [23]? (2) How might the adolescent's' voice [23] and personal agency be prioritized as they undertake the challenging task of nutritional restoration in the early stages of treatment within a treatment non-negotiable framework [50]?; (3) What is going on for the family and how might problematic family dynamics, including family conflict, be addressed? (4) At what point and how do therapists assess whether the allocation of responsibility to the parents is working and when this intervention might be contraindicated for an individual family?; (5) How might interventions more comprehensively address complex identity negotiations for adolescents and their family members? [45]; and (6) To what extent might interventions be flexibly tailored to the adolescent and family needs and preferences within a treatment non-negotiable framework [50] that prioritizes and supports the adolescent's physical and psychological safety?
Furthermore, future treatments might consider what is already becoming increasingly recognized, that family treatments for adolescent AN need to consider behavioural, psychological symptomology and family functioning in parallel and to map treatment effectiveness to these outcome measures [26].

Study strengths and limitations
The current study recruited a population of whom the majority had either dropped out and/or continued with ongoing psychological distress post treatment with FBT. Qualitative papers are, by design, in-depth investigations of small samples. However, it is usual to continue to interview or seek data until thematic saturation is reached. As the wording of the advertisement indicated inclusion of those who had discontinued FBT, this may have inadvertently excluded participants who continued to experience psychological distress post FBT suggesting that further research on this group is indicated. With this targeted recruitment of participants there exists a risk of negative bias in participants' recollection of their FBT treatment experiences in the event that participants subsequently engaged in more positively experienced treatments. On the other hand, this study goes some way in mitigating against the risk of positive bias in participant recollection of their treatment experiences that may arise in the context of a positive outcome and recovery. Another strength of the study was two author extraction of data to reduce the risk of bias in interpretation of the transcript data. Furthermore, the scope of this study is in the development of a better understanding of why FBT does work for all and why.
Negative experiences in a treatment intervention are important to understanding, however, empirical study of methods of change are required to suggest modifications to treatment that may lead to improved outcomes. It is not possible to know whether, if these adolescents' treatment preferences had been met at the time of treatment with FBT, treatment outcomes would have been improved. Also unknown were the skills and training of the FBT therapists' who delivered the individual interventions, including their capacity to build a therapeutic alliance and address the adolescent and/or family's distress.
What is known, is that the findings would suggest further research is needed to test, and find ways to test, the clinical questions that have been raised in this paper in regard to improving both treatment engagement and outcomes for adolescent AN. Future research may also seek to conduct similar research to explore the experiences and perspectives from a more diverse group of adolescents with a lived experience of adolescent AN. Further research would benefit from qualitative data triangulation (from multiple sources including adolescent, parent and clinician experiences of FBT) for completeness, convergence and dissonance of the key themes identified in this paper.

Concluding remarks
This current study highlights the complexity that is involved in the treatment of adolescent AN and consideration of systemic family issues, adolescent psychological distress and identity formation in family-based treatments. Many adolescents experience physical and psychological symptom improvement with FBT. Nevertheless, the need for greater focus on addressing psychological distress, in all phases of treatment, was identified by all the adolescents in this study as putative changes that would improve experiences and outcomes in AN treatments. Further research is needed to extend the current findings, including the tailoring of treatment to the adolescent and family needs and preferences [66] and addressing more comprehensively questions of identity, and how they may be applied in the development and evaluation of transformative AN treatment interventions.
it authentic. I eagerly await publication of the final paper. I feel it will prove instructive for my parents, going forward.
Harley: Addition to extract 2: "The FBT Program in my opinion has a narrow focus on food and weight restoration, rather than overall psychological wellbeing. For me it was harder than what it should have been because FBT unmasked other psychological issues which the anorexia had been used to deal with. The 'manualised' program approach failed to accommodate individual needs".
Amy: lt looks great though, very interesting report. Charlotte: I am happy with the paper and would be most grateful to receive a copy of the paper if published. Thank you very much for including me in the study.
(Note: Some participants were unable to be contacted).