Obstacles and facilitators of therapeutic alliance among adolescents with anorexia nervosa,their parents and their psychiatrists: A qualitative study

Abstract

Background:

Therapeutic alliance is one of the most important aspects of treatment of adolescents with anorexia nervosa. Little is known about the facilitators and obstacles of its process in this situation. We aimed to explore the experience of therapeutic alliance in inpatient treatment among adolescents with anorexia nervosa, their parents and their psychiatrists.

Methods:

This qualitative study, using semi-structured interviews, took place in France. Data collection by purposive sampling continued until we reached theoretical sufficiency. Data analysis was thematic.

Results:

Forty-one participants were included, 15 teenaged girls, 18 parents and 8 psychiatrists. Analysis showed two themes: (1) what facilitates an alliance in treatment – with four facilitators: (a) human qualities, (b) an active role in the treatment, (c) taking time and (d) taking care of the entire family and (2) what impedes an alliance in treatment with four obstacles: (a) being too close or too distant, (b) focusing on weight, (c) control and constraints and (d) psychiatrization.

Conclusion:

Collaborative work between paediatricians and psychiatrists could facilitate therapeutic alliance with parents. Definition of therapeutic alliance in this situation should be enlarged to include the adolescent–parent relationship. It is necessary to construct specific items to integrate these specific aspects to existing scales.

Keywords

Anorexia nervosa therapeutic alliance adolescents qualitative research

Introduction

Anorexia nervosa (AN) is a condition that develops most often during adolescence, characterized by a distorted body image and restricted food intake that leads to severe weight loss (Le Grange & Loeb, 2007). The mortality rate is high during this period and somatic and psychiatric complications are frequent (Steinhausen, 2009). The treatment trajectory of adolescents with AN may have a chronic or relapsing course, with episodes of dropping out of treatment. A community-based study prospectively examined the long-term outcome of 51 adolescents with AN and found that, after 18 years, 12% of the sample still had an eating disorder and 39% of them a psychiatric disorder other than an eating disorder (Wentz, Gillberg, Anckarsater, Gillberg, & Rastam, 2009). Another prospective study of adolescents with severe AN who underwent inpatient treatment found that by 10–15 years after discharge, 76% of this clinical sample had achieved full recovery but noted that 29.5% relapsed following hospital discharge(Strober, Freeman, & Morrell, 1997).

On one hand, adolescents with AN are described as resistant to the treatment, due to the ego-syntonic nature of the disease (Fassino & Abbate-Daga, 2013); on the other hand, treatment programmes that fail to be tolerable to patients result in poor adherence and treatment retention.

Medical research has extensively examined the issues related to the treatment of AN: its long-term outcomes (Strober et al., 1997; Wentz et al., 2009), the effectiveness of different psychotherapeutic approaches to it, including family-based treatment (FBT) (Lock, 2013) – that has developed evidence that justifies its recommendation in clinical practice guidelines – and enhanced cognitive behaviour therapy (Dalle Grave, Calugi, El Ghoch, Conti, & Fairburn, 2014) as well as of different treatment settings, such as day hospitals or full-time inpatient hospitalization (Herpertz-Dahlmann et al., 2014). While quantitative studies aim to evaluate outcomes in terms of reductions in scores on measures of eating disorder symptoms, qualitative studies seek to describe and understand in-depth complex phenomena associated with them. They are thus a tool of choice for focusing on the views of patients, including adolescents (Sibeoni, Costa-Drolon, Poulmarc’h, et al., 2017; Sibeoni, Orri, Podlipski, et al., 2018). We recently performed two metasyntheses, that is, systematic reviews combined with thematic syntheses, of qualitative studies. The first, reviewing 30 articles, explored how AN was experienced by adolescents, their families, and the health professionals who provide care for them and compared the perspectives of these three groups (Sibeoni, Orri, Colin, et al., 2017). The second included 32 studies and synthesized the views about treatment of these same three groups of stakeholders (Sibeoni, Orri, Valentin, et al., 2017). The results of these two metasyntheses found important relational themes that echoed the central role of the therapeutic alliance in the treatment of adolescents with AN.

Therapeutic alliance, as defined by Bordin (1993), is ‘a mutual understanding and agreement about the goals for change and the tasks necessary to advance toward these goals while awaiting the creation of bonds to support the work of all involved’. A qualitative study has shown that adolescents with AN identify the therapeutic alliance as one of the most helpful aspects of treatment (Zaitsoff, Pullmer, Menna, & Geller, 2016). A recent meta-analysis showed that early symptom improvement directly influenced the relation between therapeutic alliance and treatment outcome in eating disorders and also highlighted that alliance affected more the outcome of younger patients (Graves et al., 2017). Furthermore, several quantitative studies have emphasized the importance of establishing a therapeutic alliance in treating AN in adolescents, in terms of outcomes in hospitals and institutions (Bourion-Bedes et al., 2013), individual and family psychotherapy (Pereira, Lock, & Oggins, 2006), in particular FBT for both the adolescents and the parents (Forsberg et al., 2014; Isserlin & Couturier, 2012), and finally of dropout rates (Hubert et al., 2013).

These quantitative studies, however, are based on scales such as the Adolescent Working Alliance Inventory (AWAI) (DiGiuseppe, Linscott, & Jilton, 1996) or the Helping Alliance Questionnaire for Children and Parents (HAQ-CP) (Kermarrec, Kabuth, Bursztejn, & Guillemin, 2006), which measure specific results of alliance without approaching its process. Despite this consensus about the central role of the therapeutic alliance in adolescent AN treatment, little is known about the facilitators and obstacles of its process in this particular situation. Neither its establishment nor the factors that can facilitate or impede it have been described or studied. We found no qualitative study exploring the process of therapeutic alliance among adolescents with AN and only two qualitative studies among adults with AN. Sly et al. (2014) described the factors influencing the therapeutic alliance in inpatient treatment among women with AN: their first impressions, their active role in their treatment and a genuine dialogue with the health care professionals. Gulliksen et al. (2012) explored traits of professionals that patients consider to facilitate the therapeutic alliance and found four: acceptance, vitality, challenge and expertise. No qualitative study has crossed the perspectives of these groups of patients, families, and professionals on this topic, even though the intersection of their viewpoints enables a better understanding of their shared representations of the disease and its treatment. In recent years, we have conducted several qualitative studies exploring such crisscrossed approaches in paediatric populations (Sibeoni, Costa-Drolon, Poulmarc’h, et al., 2017; Sibeoni, Orri, Podlipski, et al., 2018).

The aim of this qualitative study was to elicit facilitators and obstacles of therapeutic alliance during a full-time inpatient treatment – that is a multidisciplinary and intensive treatment with both individual and group therapies, paediatric and psychiatric interviews, medication if needed and weight restoration process including a separation period – by exploring the experience alliance among adolescents with AN, their parents and their psychiatrists.

Crossing these three perspectives means conducting an in-depth exploration with each group of this shared experience of therapeutic alliance, not to compare but to better understand it by the integration of the views of all the stakeholders. By doing so, we might generate new insights about the issues around therapeutic alliance in this particular situation that might have new clinical implications and potentially improve the care process.

Methods

Setting

This exploratory multicentre study took place at the Department of Adolescent Psychiatry at the Argenteuil Hospital Centre, in the outpatient unit receiving adolescents with AN who had previously been hospitalized in specialized departments in the Ile de France region, and at the Rouen University Hospital Centre, the regional referral centre for the most severe cases of AN, those not responding to first-line treatment by their local mental-health professionals. Both departments prefer outpatient management after a short hospitalization focused on starting the weight restoration process, with psychiatric support. If either of these two types of management fails, full-time hospitalization in a specialized unit is scheduled, adapted case-by-case over a duration of several weeks. All the adolescents recruited in this study had undergone a full-time inpatient treatment in such specialized unit. Their hospitalization involved a total separation, as part of the weight restoration process, at the beginning that never exceeded 10 weeks and was adapted on on a case-by-case basis. They all received individual psychotherapy, and different types of expressive and other therapy with psychomotor and occupational therapists, and some received pharmacological treatment. This inpatient treatment was multidisciplinary (paediatrician, psychiatrist, psychologist, dietitian, physical therapists, expressive and creative arts therapists, social worker, as well as classroom time for school subjects and field trips) and planned to include the parents, the adolescents, and the siblings. Both adolescents and parents could participate in support groups.

The council for the evaluation of ethics in health research of the University of Paris–Descartes approved the protocol. All adolescents, their parents and their psychiatrists provided written consent before inclusion in the study. This study complied with the COREQ (COnsolidated criteria for REporting Qualitative research) guidelines (Tong, Sainsbury, & Craig, 2007).

Sampling and participants

Sampling was purposive (Gentles, Charles, Ploeg, & McKibbon, 2015) that is selective and deliberate. The researchers explained the study design in detail to the physicians collaborating in the study. The latter then identified the potential participants (adolescents and parents) whom they considered likely to provide the most information.

We established a list of inclusion and exclusion criteria for the participants (Table 1). Clinical coordinators in each department, all experienced psychiatrists (M.-A.P., P.G., E.J.), ensured adherence to the inclusion and sampling criteria. They assessed potential participants during a clinical interview to verify they meet The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for AN as the main diagnosis.

Table 1.

Inclusion and exclusion criteria.

Inclusion criteria	- Age: 12–18 years - Diagnosis of anorexia (according to DSM-5) - Speaks French fluently - In treatment for more than 6 months - Hospitalized for at least 1 month - AN as the principal diagnosis and initial reason for care - Interested in participating in the study, with parents and psychiatrist
Exclusion criteria	- Somatic comorbidities - Psychiatric comorbidities, such as schizophrenia, bipolar disorder or autism spectrum disorder - Intellectual disability - Acute symptoms at inclusion

DSM-5: The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; AN: anorexia nervosa.

All forms of AN (pure restrictive AN, or with bingeing and/or purging behaviour) were included because of the frequent shifts between these different types. The psychiatrists interviewed were the referring physicians of the adolescents; they were all involved in the inpatient treatment of their patients, some directly by working in the unit and others by having close contact and participating in all the medical decisions.

The collaborating physicians first mentioned the study to potential participants and gave them an information sheet about it. If they were interested, one of the researchers spoke with the adolescent and his or her parents in a preliminary meeting to present the study to them, obtain their written consent and collect social and demographic data.

Data collection

Data came from semi-structured interviews with an open-ended approach (Britten, 2013). The interviewers used an interactive conversational style and a list of areas of experience to be explored with adolescents, parents and psychiatrists (Table 2). The participants were considered to be the experts on their own experience, and the interviews were conducted in a way that offered them the opportunity to recount it.

Table 2.

Interview guide.

Areas of experience	Potential questions with an adolescent
Disease history, onset of the disease	How long have you had this disorder? How was it discovered?
Help-seeking process	Can you tell me about how and when you started receiving psychiatric care?
The beginning of treatment	Can you tell me about your first meeting with your psychiatrist and with the other staff?
Treatment decisions	How did you understand and experience the treatment plan?
Therapeutic relationship	At first, what did you think about the relationship between you and the psychiatrist and the other staff
Treatment options and decision sharing	What treatment options were suggested to you? How the decision was made?
Therapeutic alliance	How was the alliance between you and the staff? Between your parents and them?

The objective of the interview was to obtain an in-depth description by participants of their experience of the early therapeutic alliance in all its aspects.

A single researcher interviewed the adolescent, then one or two parents and then the referring psychiatrist. There were therefore three interviews related to each teen with AN. Each lasted from 60 to 90 minutes. They were conducted by two researchers (J.S. and L.P., psychiatrists treating adolescents and M.O., a clinical psychologist), from September 2015 through March 2016, and then from February through June 2017. The interviews, anonymized, were recorded and transcribed verbatim. The transcript thus obtained was then analysed, as described below.

Our sample size was determined according to data saturation following the principle of theoretical sufficiency (Dey, 1999): data collection and analysis were completed when the researchers determined that the themes obtained offered a sufficient explanatory framework for the data collected. In practice, data analysis took place simultaneously with the data collection and inclusion continued for as long as the analysis of the material continued to provide new information that would altered the themes or subthemes. Theoretical sufficiency was considered reach after 39 interviews, when the analysis of new data no longer changed the thematic organization of our results. Interviews related to two other teens (six interviews in all) were performed without producing new themes, to ensure this data saturation.

Analysis

Thematic analysis was used to explore the data (Braun, Clarke, & Terry, 2014). This method enables the identification, analysis and reporting of the themes within the data. Our thematic analysis used an inductive approach, coding the data without any reference to theoretical notions or the researchers’ preconceptions. Table 3 summarizes the different stages of our thematic analysis. This process was dynamic and iterative, with each new transcript leading to the collection of additional data and to their subsequent analysis. The objective was to identify the similarities and the differences between the accounts of each participant. The researchers were thus led to discern the recurrent patterns but also to integrate the new issues that emerged from the analysis. Three researchers (J.S., L.V. and A.R.-L.) independently performed this analysis with NVivo 11 software. Moreover, results were regularly debated during research group meetings. In cases of disagreements, the discussion continued until a consensus was reached.

Table 3.

Process of thematic analysis.

	Activities	Rationale
Stage 1	Repeatedly read each transcript, as a whole	Obtain a global picture of the interview and become familiar with the interviewee’s verbal style and vocabulary. Each new reading of the transcript might also provide new perspectives.
Stage 2	Code the transcript by making notes corresponding to the fundamental units of meanings.	Make descriptive notes using the participant’s own words.
Stage 3	Make conceptual notes through processes of condensation, abstraction and comparison of the initial notes.	Categorize initial notes and reach a higher level of abstraction.
Stage 4	Identify initial themes. Provide quotations that illustrate the main ideas of each theme.	Themes are labels that summarize the essence of a number of related conceptual notes. They are used to capture the experience of the phenomenon under study.
Stage 5	Identify recurrent themes across transcripts and produce a coherent ordered table of the themes and subthemes.	Move from the particular to the shared across multiple experiences. Recurrent themes reflect a shared understanding of the phenomena among all participants. During this more analytic stage, researchers try to make sense of the associations between the themes found.

Results

The study included 15 teenaged girls, 18 parents (14 mothers and 4 fathers) and 8 psychiatrists (trainees, chief residents and hospital staff physicians, 6 women and 2 men, aged from 27 to 50 years) for a total of 41 participants (Table 4). Each physician was interviewed separately for each participating patient, on average, twice (range: 1–4). All families were French citizens. In all, 45 interviews were performed, with five couples of parents interviewed together.

Table 4.

Characteristics of the adolescents and parents.

	Adolescents										Parents
	Psychiatrist interviewed	Age	Duration of hospitalization (months)	Subtype of AN	Treatment received during FTH	Comorbidity	BMI (kg/m²) (admission)	BMI (kg/m²) (discharge)	BMI (kg/m²) 6 months	BMI (kg/m²) 1 year	Parents	Interviewees	Relationship situation	Profession
A1	MD1	16	5	R	IP, SSRI	Depression	15.5	20.6	20.1	23.4	P1	Mother and father	Couple	M: teacher F: teacher
A2	MD2	15	5	R	IP, SSRI	Depression	13.4	17.9	NA	NA	P2	Mother	Couple	M: executive F: horticulturist
A3	MD2	17	8	B/P	IP, SSRI + AP	Anxiety	14	17.4	18.2	18.7	P3	Mother	Couple	M: physician F: banker
A4	MD3	14	4.5	R	IP, FT, SSRI	Depression	18	18.1	20.6	21.4	P4	Mother	Separated	M: nurse F: executive
A5	MD2	15	8.5	B/P	IP, SSRI	Depression	14.5	17.2	15.5	19.1	P5	Mother and father	Couple	M: social worker F: journalist
A6	MD4	14	3	R	IP	None	12.9	17.3	17.1	19.0	P6	Mother	Couple	M: pharmacy technician F: salesman
A7	MD2	17	8	R	IP, SSRI + AP	Depression	15.1	18	16.9	16.5	P7	Mother	Couple	M: health executive F: orthoptist
A8	MD5	13	2	R	IP, FT, SSRI	Depression	16,9	20.2	22.0	26.8	P8	Mother	Couple	M: teacher F: teacher
A9	MD1	14	4	R	IP, SSRI	None	14.5	16.4	15.1	15.9	P9	Mother	Couple	M: nutritionist F: engineer
A10	MD1	14	4	R	IP, SSRI + AP	Depression	15.5	19.1	17.7	16.6	P10	Mother and father	Couple	M: teacher F: teacher
A11	MD4	16	5	B/P	IP	None	14.5	19.1	17.3	17.5	P11	Father	Couple	M: lawyer F: lawyer
A12	MD6	15	6	R	IP, FT, SSRI	Anxiety	14.9	19	18.1	18.1	P12	Mother	Separated	M: executive F: paralegal
A13	MD7	16	7	B/P	IP, SSRI	Depression	13.7	17.8	17.7	17.7	P13	Mother and father	Couple	M: CEO F: unemployed
A14	MD8	15	4.5	B/P	IP, AP	Depression	15	20.1	19.3	18.9	P14	Mother and father	Couple	M: Police officer F: administrative
A15	MD6	17	8	R	IP, FT, SSRI	None	15.1	19.4	19.6	20.1	P15	Mother	Separated	M: NA F: sports teacher

AN: anorexia nervosa; SSRI: selective serotonin reuptake inhibitor; BMI: body mass index; AP: antipsychotic; R: restrictive; B/P bingeing and/or purging; IP: individualized psychotherapy; FT: family therapy; FTH: full time hospitalization.

All of the adolescents recruited agreed to participate. All had been discharged from an inpatient admission during the year before the interview. Their ages ranged from 13 to 17 years and averaged 15.2 years. Their cumulative length of stay ranged from 2 to 8.5 months, with a mean of 5.5 months. Thirteen had received pharmacological treatment: selective serotonin reuptake inhibitors only (SSRI) (N = 9), SSRI and atypical antipsychotics (N = 3), or atypical antipsychotics only (N = 1). Eleven received this treatment because of a comorbid condition, either a depressive disorder (N = 9) or an anxiety disorder (N = 2) or anxiety symptoms (N = 2). We chose to organize our thematic analysis around the two issues of our research question: (1) what facilitates an alliance in treatment and (2) what impedes an alliance in treatment. The results are presented below and relevant quotations (from the interview transcripts and translated from French into English for the purpose of this article) in Table 5.

Table 5.

Direct quotations from interviews.

1. What facilitates an alliance in treatment

1.1. Human qualities and relational skills of health care providers

Adolescents

Q1 – A2: She was the first one I saw when I arrived, she seemed very gentle and she reassured me.

Q2 – A7: The nurses were so close to me that I had the impression that they’d replaced my family.

Physicians

Q3 – MD8: I found that this family was really agreeable to work with . . . because one could really develop things, say things.

Q4 – MD7: Yes, more motherly and more understanding . . . even when she was setting limits, which probably reinforced the parental aspect.

Parents

Q5 – P14: The team, they weren’t judgmental, that is super important also, they led things very subtly, finally.

1.2. Playing an active role in the treatment

Adolescents

Q6 – A14 – To the question ‘What helped you?’: The fact that I could count on my parents even though we might fight a lot.

Physicians

Q7 – MD4: ‘We don’t know your teenagers very well, we’re listening to you’. That was the idea.

Parents

Q8 – P2: What I really appreciated, it’s that already, things weren’t done, they talked to us about it first, each time.

1.3. Taking time

Physicians

Q9 – MD4: ‘It’s starting over!’, and so she rushed here. ‘Can I come back?’ I told her ‘of course you can come back’. So now she comes almost every day.

Parents

Q10 – P15: As a parent, I have the impression of a net – she won’t fall anymore where she was, she’d be caught, as by a transparent net at the circus.

Adolescents

Q11 – A6: off course, the more you know them, the easier it gets [. . .] It’s like you need time to trust someone.

1.4. Taking care of the entire family and improving family relationships

Adolescents

Q12 – A7: My mother gave her a cup when she left, I think that proves how much she was able to help my parents.

Q13 – A1: . . . to re-establish relationships with my family, especially my sister, my parents, relationships that are calmer. This, it’s all concrete things [that helped me to get better].

Q14 – A5: It’s as if, being with each other, we maintain the pain. And then as a result, the separation lets us clarify things. I thought about myself, they thought about themselves, and we found each other again in better conditions.

Physicians

Q15 – MD6: And she took a ridiculous amount of time before she talked to us, it took almost two and a half months for her to talk to us. But one day, she was able to talk to us about it, so we were able to free ourselves to talk about eating disorders.

Parents

Q16 – P4: I dared to ask all the questions, I dared to let go of all my fears, my trepidations.

Q17 – P12: I had to manage three women, it was horrendous, so I found this solution and we left for the ER in Aix. [. . .] But the aim of the manoeuvre was to go to the ER, but it was almost an incidental aim. The aim, it was already to take a break, to go as slowly as possible by automobile, because it was a moment, a period of decompression.

Q18 – P13: The first time they put your baby in your arms, as the father, you tell yourself, ‘that’s what I needed . . . I have a job to do, to raise a child’.

2. What impedes an alliance in treatment

2.1. Being too close or too distant

Adolescents

Q19 – A6: I wasn’t at all comfortable and I really had the impression of being interrogated. And I didn’t like that at all.

Parents

Q20 – P5: When we’re obviously talking about our personal lives, private stuff, family [. . .] And to have all those people looking and listening, sometimes, it’s hard.

Q21 – F13: Both the male and the female nurses, they were open with the young people. They did stuff, we said, ‘they’re exaggerating!’

2.2. Focusing on weight gain

Adolescents

Q22 – A6: They saw us as all the same, because we were thin, underweight. And, in fact, it’s not that, it happens mostly in the head. It’s that there’s something that’s wrong, we each have a different story.

Q23 – A15: He put me in a general category . . . he said, ‘yes anorexic people they do this, they don’t do that’ and I had the impression that he wasn’t treating me as a patient but really as the disease.

Physicians

Q24 – MD8: . . . that she really weighed much too little and that she also needed to gain weight and relearn how to eat to not aggravate the symptoms, and she was in a really difficult physical state.

Q25 – MD3: There was really the mask in place, ‘no, I’m going to eat, everything is fine’, and when she was transferred to the 4th floor and she understood that the hospitalization would be longer [. . .], she let the barriers down a little.

Parents

Q26 – P2: A recognition of her disease, even though she wasn’t really yet fully aware of it, but to have put it into words, to have heard, ‘yes, you are really sick’.

2.3. Control and constraints

Adolescents

Q27 – A6: Because they already imposed a lot on me. For me, it was really too much, they were imposing too many things all at once. It was going too fast. And I had an argument at the appointment with the child psychiatrist.

Parents

Q28 – P6: You still tell yourself that you don’t have much choice, because in any case you don’t know what to do any more. You have your back to the wall and have to say, ‘There’s nothing else to do’.

Physicians

Q29 – MD5: I only do it with the teen’s consent, if the teen agrees for me to see her parents alone.

2.4. Psychiatrization

Adolescents

Q30 – A1: I think that third parties, they tend to take it personally, therefore ‘this anorexic behaviour is directed against me’. And as a result often they tend to enter into a form of conflict or rejection.

Physicians

Q31 – MD8: The hospitalization side of a department of psychopathology, for them, in their heads . . . they really experience it as a failure.

Q32 – MD3: She stayed very much on the very very medical side of things, we didn’t succeed in talking about much else.

Q33 – MD5: They were more or less in denial about what was happening to their daughter, a little normalization, had trouble assessing not only what was going on, but also what was waiting for them.

What facilitates an alliance in treatment

Participants described four facilitators of the therapeutic alliance: (1) human qualities and relational skills of health care providers, (2) playing an active role in the treatment, (3) taking time and (4) taking care of the entire family and improving family relationships.

1. Human qualities and relational skills of health care providers

All the participants considered that human qualities and relational skills of the hospital staff generally, and the physician in particular, facilitated the establishment of a therapeutic alliance.

The adolescents were very sensitive and attentive to the human qualities of the professionals around them, especially their reliability (Q1) and their authentic involvement in the teen’s management. They emphasized the importance of feeling listened and understood by the staff and especially by the specialists in eating disorders, with whom they could talk about their disorder without fear of judgement. The adolescents also insisted on the relational skills of the professionals. They considered that their interactions with the hospital staff should be comparable in intensity and quality to that in family relationships. They mentioned the family-like and maternal aspects of the staff as a positive element for building a good alliance (Q2).

The physicians also underlined the importance of the human qualities and of good feelings in their relationships with both the teens and their parents. They insisted particularly on the relational stakes of the first meeting (Q3). Doctors underlined the importance of a solid bond, resistant to difficulties. The physicians considered that the adolescents, while hospitalized, should experiment with calmer, more understanding interactions with adults while carefully maintaining an appropriate relational distance (Q4).

Parents also underlined the importance of the staff’s human and relational qualities, especially their kindness (Q5)

2. Playing an active role in the treatment

Adolescents explained explicitly that playing an active role in their care was a key to a good therapeutic alliance. They also mentioned the importance of the parents’ involvement in the treatment. That is, the adolescents considered that the fact that their parents appeared concerned and mobilized around them was an essential factor in promoting the therapeutic alliance (Q6).

Doctors recognized the importance of the adolescent active role do facilitate alliance and sought to underline the value of the parents’ role in treatment and to consider them as ‘experts’ about their daughters (Q7).

Mirroring the doctors’ attitudes, parents were especially sensitive to the doctors’ frankness towards them and systematic inclusion of them in decision making (Q8).

3. Taking time

All participants mentioned the link between time and therapeutic alliance.

Doctors mentioned the time necessary to establish an association as solid as possible as the basis of the alliance that makes treatment possible (Q9). They explained that facilitating alliance necessitates taking the time for explanations and respecting the parents’ rhythm, especially by leaving them the possibility of interrupting the treatment, if they did not feel ready. The psychiatrists also explained the need not to rush the parents and to respect their defences around sensitive or painful subjects.

Parents valued the team’s investment over time in the adolescents and their families (Q10).

Teens recognized the value of time to establish therapeutic alliance; that time was necessary to trust to the staff and better adhere to treatment (Q11). However, they also complained about hospitalizations lasted too long and considered this as an obstacle to maintain therapeutic alliance in the long run.

4. Taking care of the entire family and improving family relationships

Taking care of the entire family was considered as a facilitator by all participants. They particularly all mentioned the adolescent–parent relationship as the central element of the therapeutic alliance.

The teens considered that their parents too suffered and that they needed support during their daughters’ hospitalization (Q12). They pointed out the importance of providing care for the distress of the entire family, including siblings. They felt reassured and confident when this overall management was set up. They also insisted on the importance of improving family relationships (Q13). According to them, the change experienced in the parent–adolescent relationship during treatment was a key to establish and maintain therapeutic alliance. In that matter, adolescents mentioned the calming function of the separation because of the hospitalization (Q14).

The physicians considered that the treatment team’s management of the parents’ distress, by supporting them in their daily lives and in welcoming their life stories, is essential, a supplement to the teens’ treatment that enables the construction of a good therapeutic alliance (Q15). The physicians also perceived the improvement in the adolescent–parent relationship and its effect on the therapeutic alliance. They explicitly described the appearance of a new view of eating disorders and the establishment of a new type of relationship on the parents’ side, which served to facilitate the therapeutic alliance.

Parents too recognized their own distress and felt that they needed support (Q16). They did not think that a therapeutic alliance could exist without providing care for the entire family. They wanted treatment to result in relieving family tensions (Q17). As a corollary, some parents reported that the prolonged absence of the hospitalized daughter might cause family divisions and felt that the staff did not pay enough attention to the family and especially to the siblings. Calming of the tensions in the adolescent–parent relationship promoted the parents’ alliance in the treatment (Q18). Parents thus reported the progressive construction of new ways of relating to their daughters, which appeared to them to be better adapted to the child’s condition and enabled higher quality interactions.

What impedes an alliance in treatment

Participants described four obstacles of the therapeutic alliance: (1) being too close or too distant, (2) focusing on weight, (3) control and constraints and (4) psychiatrization.

1. Being too close or too distant

The adolescents perceived the attention of staff members as excessive or overly protective. They said that interviews were sometimes too private and personal and thus annoying (Q19). This feeling of intrusion was an obstacle to establish and maintain therapeutic alliance.

Parents also revealed a feeling of intrusion, of being ‘dissected’ or ‘patted down’ during interviews. They said that it could make them ill at ease and engender distrust or mistrust towards the physician (Q20). They also expressed concerns about the relation distance, judging the positioning of the staff as either too close and inappropriate or too distant (Q21).

2. Focusing on weight gain

Focusing or organizing treatment for the sole purpose of regaining weight was perceived by the adolescents as an obstacle to the therapeutic alliance (Q22). Moreover, adolescents explicitly expressed their need to be considered as individuals and not diseases. The absence of a personalization of treatment led them to feel categorized and stigmatized as ‘anorexic’, which impaired the therapeutic alliance (Q23).

Parents and physicians, however, described weight gain as an essential objective (Q24). The physicians perceived that disagreement about this objective between the adolescents and themselves undermined the therapeutic alliance (Q25). Many parents considered that the primary objective of care was to make their daughters aware of the danger of this eating disorder, without which no quality therapeutic alliance could exist, and to obtain her agreement to the objective of gaining weight (Q26).

3. Control and constraints

All three groups raised the issues of control, constraints and balance of power between the adolescents, the parents and the physicians. The girls considered that their exclusion from decisions and the constrained imposition of treatment made any therapeutic alliance impossible (Q27). Parents also reported perceiving constraint and an absence of choice in the means of treatment proposed (Q28) and described an asymmetric relationship and a balance of power with the physicians that were not propitious to the development of a therapeutic alliance. The physicians, however, reported feeling that they were working in collaboration with the patient and her family (Q29).

4. Psychiatrization

Finally, the issue of AN perceived as a psychiatric disorder requiring psychiatric treatment directly hindered the alliance.

The adolescents reported that the staff did not understand eating disorders. They complained that the personnel reduced their disorders to a list of symptoms or understood them as an attitude either hostile or aggressive (Q30). The physicians reported that parents often experienced psychiatric management as a failure and doubts about their parenting skills (Q31). Similarly, they observed that it is difficult to ally themselves with parents who have a negative representation of psychiatry and considered the somatic aspects of management most important (Q32). They also noted that parents seemed to minimize the psychiatric severity of their children’s disorders, judging the psychiatrists’ discourse alarmist and exaggerated (Q33).

Discussion

Everyone knows and recognizes the stakes of the therapeutic alliance in the treatment of AN in adolescents. Nonetheless, no study has ever explored the process – what happens between the adolescent, his or her parents and the hospital staff to construct this alliance.

Our results identify facilitators already known and described in the literature. Numerous aspects of our results echo items in the scales most frequently used to assess therapeutic alliances (DiGiuseppe et al., 1996; Kermarrec et al., 2006): comfort in the therapeutic relationship, feeling understood, staff respect and involvement, shared decision making and so on. These aspects found in scales to measure therapeutic alliance are not specific to the alliance for the treatment of AN in adolescents.

The impediments to the alliance process found in our results are linked to specific issues already described, in particular the ego-syntonic nature of eating disorders and the need for adolescents to be aware of them, the disagreements around the priority of the objective of gaining weight and the issues of control in treatment (Sibeoni, Orri, Valentin, et al., 2017).

Two points appear important to discuss.

Our results found that parents’ negative representations of psychiatry and favouring somatic aspects of treatment impede the establishment of an alliance. Generally, regardless of the psychiatric disease, prejudices and stereotypes about psychiatry can have a negative effect on care and can hinder the adherence of adolescents and their parents (Moses, 2010). Depending on the specific country and the specific organization of care, outpatient and inpatient treatment of AN in adolescents can take place in adolescent psychiatry units, which were the context for our study, or in units specialized in the treatment of eating disorders, or even in departments of paediatric and adolescent medicine. AN is a psychiatric disorder that also requires somatic management. The linkage of somatic and psychiatric treatment appears here to be an issue affecting the therapeutic alliance with parents. Collaborative work between paediatricians and psychiatrists should make it possible to avoid reinforcing parents’ negative representations of psychiatry while meeting their expectations for somatic management. This is the aim of the model of the ‘Maisons des adolescents’ in France, facilities that connect primary somatic and psychiatric care and thus help to destigmatize the image of psychiatry (Benoit, Smadja, Benyamin, & Moro, 2011).

We expected to find at least a relational dimension to be a factor related to the therapeutic alliance, as previously described in our metasynthesis (Sibeoni, Orri, Colin, et al., 2017; Sibeoni, Orri, Valentin, et al., 2017) but also in numerous empirical and theoretical works about relational and attachment issues in eating disorders generally and in AN in particular (Jewell et al., 2016). The question of relationship is in fact everywhere in our results. Not only do we find that the quality of the relationships between adolescents and staff and between parents and staff is important, but also the family relationships and the adolescent–parent relationship in particular. Our results suggest that the adolescent–parent relationship may play a key role in the construction of the therapeutic alliance for the treatment of AN in adolescence. We thus consider it necessary to enlarge the definition of the therapeutic alliance in this situation so that it includes as an integral part the adolescent–parent relationship.

Similarly, in the scales measuring the therapeutic alliance used in this population, such as the AWAI or the HAQ-CP, the items about relationships concern only the relationship with care providers and focus on its perceived quality (e.g. in the AWAI: ‘The therapist and I like each other, The therapist and I respect each other, The therapist and I understand each other, I feel uncomfortable with the therapist, I feel that the therapist appreciates me, my relationship with the therapist is very important to me, I feel the therapist cares about me even when I do things that he/she doesn’t approve of’).

In our results, however, relationships are involved in all three components of the definition of the therapeutic alliance – the quality of the association, the objective of treatment and the means to achieve it – but none of the scales concerns the relationship as either an objective or means of treatment. Similarly, they do not mention the role of the adolescent–parent relationship. It may thus be relevant to add some specific items to assessing the therapeutic alliance in the treatment of AN in adolescents.

We therefore propose, on the basis of our results, to construct specific items around (1) the quality of the relationship with staff as a means of getting better, (2) the impact on the alliance of parental involvement in treatment, (3) the impact of treatment on the parents’ point of view about the adolescent and the relationship between adolescents and their parents and finally (4) the agreement about the objectives for improving family relationships generally and adolescent–parent relationships in particular.

Finally, it would be interesting to explore with the same study design the experience of alliance of FBT (Lock, 2013). The role of therapeutic alliance in FBT for adolescent AN is essential and both the parents–therapist and the adolescent–therapists alliance have been studied underlining how important the quality of the relationship with the therapist and the involvement of the parents are (Forsberg et al., 2014; Isserlin & Couturier, 2012). To our knowledge, the role of adolescent–parent relationships or the issue of psychiatrization has never been explored.

Strengths and limitations

This is the first qualitative study to look at the process of therapeutic alliance within the treatment of adolescent AN.

Nonetheless, some limitations must be taken into consideration. First, it took place in France, and caution is required in transposing our results to other places because adolescent treatment depends strongly on the organization of the health care system as well as on the country’s economy. Second, the population of adolescents was recruited in specialized inpatient departments of adolescent psychiatry. Our results therefore apply to this population receiving this kind of care in these types of settings, but they cannot take into account other contexts of care for the same disorder, especially outpatient-only treatment or treatment in a paediatric unit.

Third, this study’s objective was to understand the global experience of therapeutic alliance process and not to assess if adolescents, parents and psychiatrists were rating the quality of the therapeutic alliance the same or not. However, developing specific items based on our results could be useful for further research, to measure the quality of the therapeutic alliance process in general and to compare how it is perceived differently by the adolescents, the parents and the professionals. Moreover, more than half of the adolescents included had comorbidity with depression and this disorder might have impacted the way they recounted their experience regarding therapeutic alliance.

Fathers are underrepresented in this study since only four were included. However, we did not find differences between fathers and mothers in our sample of parents. Finally, we included only girls with AN in our study. Thus, our results may not be relevant for the all gender spectrum in which AN can occur.

Footnotes

Acknowledgements

We would like to thank the participants of this study, and JA Cahn for the translation.

Author’s note

Laurence Verneuil is also affiliated with Department of Dermatology, CHRU Caen, Caen Basse Normandie University, Caen, France.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the French Ministry of Health through the Programme de recherche sur la performance du système des soins (PREPS EVAADO, PREPS-15-0024).

ORCID iD

Jordan Sibeoni

Author biographies

Jordan Sibeoni is a consultant child and adolescent psychiatrist working in SUPADO, Argenteuil hospital a university adolescent psychiatry department. He has completed a Ph.D in Public Health at the University of Paris Sud regarding the development of qualitative methods in adolescent psychiatry and published a number of articles in the field using qualitative methods.

Laurence Verneuil is a professor of Dermatology. After conducting basic research in oncogenesis, she has developed an expertise in qualitative health research.

Lea Poulmarc’h is a consultant child and adolescent psychiatrist working in SUPADO.

Massimiliano Orri is a researcher psychologist, post-doc researcher in Mc Gill University, Canada. He has previously completed a Ph.D in Psychology at the University of Paris Descartes.

Elise Jean is a consultant child and adolescent psychiatrist working in SUPADO. She has previously completed a Ph.d in Biology.

Marc-Antoine Podlipski is a consultant child and adolescent psychiatrist, head of a day hospital in the Rouen child and adolescent psychiatry department.

Priscille Gérardin is a professor of child and adolescent psychiatry, head of the Rouen child and adolescent psychiatry department.

Anne revah-Lévy is a professor of child and adolescent psychiatry, head of SUPADO. Pr Revah-Levy is a major actor of qualitative research for 15 years. She is the director of the qualitative health research team.

References

Benoit

J.-P.

Smadja

Benyamin

Moro

M.-R.

(2011). Construire une relation de soins avec les adolescents. Qu’apporte le nouveau dispositif des Maisons des adolescents? Neuropsychiatrie de l’Enfance et de l’Adolescence, 59(2), 86–93.

Bordin

(1993). Theory and research on the therapeutic working alliance: New directions. In Horvath

A. O.

Greenberg

L. S.

(Eds.), The working alliance: Theory, research and practice (pp. 13–37). New York, NY: John Wiley.

Bourion-Bedes

Baumann

Kermarrec

Ligier

Feillet

Bonnemains

. . . Kabuth

(2013). Prognostic value of early therapeutic alliance in weight recovery: A prospective cohort of 108 adolescents with anorexia nervosa. Journal of Adolescent Health, 52, 344–350.

Braun

Clarke

Terry

(2014). Chapter 7: Thematic analysis. In Rohleder

Lyons

A. C.

(Eds.), Qualitative research in clinical and health psychology (pp. 95–113). Basingstoke, UK: Palgrave Macmillan.

Britten

(2013). Chapter 2: Qualitative interviews. In Pope

Mays

(Eds.), Qualitative research in health care (3rd ed.) (pp. 12–20). Malden, MA: Blackwell Publications/BMJ Books.

Dalle Grave

Calugi

El Ghoch

Conti

Fairburn

C. G.

(2014). Inpatient cognitive behavior therapy for adolescents with anorexia nervosa: Immediate and longer-term effects. Frontiers in Psychiatry, 5, 14.

Dey

(1999). Grounding grounded theory: Guidelines for qualitative inquiry. San Diego, CA: Academic Press.

DiGiuseppe

Linscott

Jilton

(1996). Developing the thrapeutic alliance in child: Adolescent psychotherapy. Applied and Preventive Psychology, 5. 85–100.

Fassino

Abbate-Daga

(2013). Resistance to treatment in eating disorders: A critical challenge. BMC Psychiatry, 13(1), 282.

10.

Forsberg

LoTempio

Bryson

Fitzpatrick

K. K.

Le Grange

Lock

(2014). Parent-therapist alliance in Family-Based Treatment for adolescents with anorexia nervosa. European Eating Disorder Review, 22(1), 53–58.

11.

Gentles

S. J.

Charles

Ploeg

McKibbon

(2015). Sampling in qualitative research: Insights from an overview of the methods literature. The Qualitative Report, 20, 1772–1789.

12.

Graves

T. A.

Tabri

Thompson-Brenner

Franko

D. L.

Eddy

K. T.

. . . Isserlin

(2017). A meta-analysis of the relation between therapeutic alliance and treatment outcome in eating disorders. International Journal of Eating Disorders, 50, 323–340.

13.

Gulliksen

K. S.

Espeset

E. M. S.

Nordbø

R. H. S.

Skårderud

Geller

Holte

(2012). Preferred therapist characteristics in treatment of anorexia nervosa: The patient’s perspective. International Journal of Eating Disorders, 45, 932–941.

14.

Herpertz-Dahlmann

Schwarte

Krei

Egberts

Warnke

Wewetzer

Dempfle

(2014). Day-patient treatment after short inpatient care versus continued inpatient treatment in adolescents with anorexia nervosa (ANDI): A multicentre, randomised, open-label, non-inferiority trial. The Lancet, 383, 1222–1229.

15.

Hubert

Pioggiosi

Huas

Wallier

Maria

A.-S.

Apfel

Godart

(2013). Drop-out from adolescent and young adult inpatient treatment for anorexia nervosa. Psychiatry Research, 209, 632–637.

16.

Isserlin

Couturier

(2012). Therapeutic alliance and family-based treatment for adolescents with anorexia nervosa. Psychotherapy, 49(1), 46–51.

17.

Jewell

Collyer

Gardner

Tchanturia

Simic

Fonagy

Eisler

(2016). Attachment and mentalization and their association with child and adolescent eating pathology: A systematic review: Attachment and mentalization. International Journal of Eating Disorders, 49, 354–373.

18.

Kermarrec

Kabuth

Bursztejn

Guillemin

(2006). French adaptation and validation of the helping alliance questionnaires for child, parents, and therapist. The Canadian Journal of Psychiatry, 51, 913–922.

19.

Le Grange

Loeb

K. L.

(2007). Early identification and treatment of eating disorders: Prodrome to syndrome. Early Intervention in Psychiatry, 1(1), 27–39.

20.

Lock

(2013). Treatment manual for anorexia nervosa: A family-based approach (2nd ed.). New York, NY: Guilford Press.

21.

Moses

(2010). Being treated differently: Stigma experiences with family, peers, and school staff among adolescents with mental health disorders. Social Science & Medicine, 70, 985–993.

22.

Pereira

Lock

Oggins

(2006). Role of therapeutic alliance in family therapy for adolescent anorexia nervosa. International Journal of Eating Disorders, 39, 677–684.

23.

Sibeoni

Costa-Drolon

Poulmarc’h

Colin

Valentin

Pradère

Revah-Levy

(2017). Photo-elicitation with adolescents in qualitative research: An example of its use in exploring family interactions in adolescent psychiatry. Child and Adolescent Psychiatry and Mental Health, 11, 49.

24.

Sibeoni

Orri

Colin

Valentin

Pradère

Revah-Levy

(2017). The lived experience of anorexia nervosa in adolescence, comparison of the points of view of adolescents, parents, and professionals: A metasynthesis. International Journal of Nursing Studies, 65, 25–34.

25.

Sibeoni

Orri

Podlipski

M.-A.

Labey

Campredon

Gerardin

Revah-Levy

(2018). The experience of psychiatric care of adolescents with anxiety-based school refusal and of their parents: A qualitative study. Journal of the Canadian Academy of Child and Adolescent Psychiatry = Journal De l’Academie Canadienne De Psychiatrie De L’enfant Et De L’adolescent, 27(1), 39–49.

26.

Sibeoni

Orri

Valentin

Podlipski

M.-A.

Colin

Pradere

Revah-Levy

(2017). Metasynthesis of the views about treatment of anorexia nervosa in adolescents: Perspectives of adolescents, parents, and professionals. PLoS ONE, 12(1), e0169493.

27.

Sly

Morgan

J. F.

Mountford

V. A.

Sawer

Evans

Lacey

J. H.

(2014). Rules of engagement: Qualitative experiences of therapeutic alliance when receiving in-patient treatment for anorexia nervosa. Eating Disorders, 22, 233–243.

28.

Steinhausen

H.-C.

(2009). Outcome of eating disorders. Child and Adolescent Psychiatric Clinics of North America, 18, 225–242.

29.

Strober

Freeman

Morrell

(1997). The long-term course of severe anorexia nervosa in adolescents: Survival analysis of recovery, relapse, and outcome predictors over 10–15 years in a prospective study. The International Journal of Eating Disorders, 22, 339–360.

30.

Tong

Sainsbury

Craig

(2007). Consolidated criteria for reporting qualitative research (COREQ): A 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care, 19, 349–357.

31.

Wentz

Gillberg

I. C.

Anckarsater

Gillberg

Rastam

(2009). Adolescent-onset anorexia nervosa: 18-year outcome. The British Journal of Psychiatry, 194, 168–174.

32.

Zaitsoff

Pullmer

Menna

Geller

(2016). A qualitative analysis of aspects of treatment that adolescents with anorexia identify as helpful. Psychiatry Research, 238, 251–256.