The goal of this experimental interventional study is to learn if the psychological treatment Radically Open Dialectical Behavior Therapy (RO-DBT), is feasible to conduct and provide positive effects for adolescents with eating disorders and emotional overcontrol. Participants will be those at risk for developing severe and enduring eating disorder symptoms, as they have not responded fully to previous treatment attempts.

Primary outcomes are feasibility and changes in eating disorder symptoms. Secondary outcomes are changes in unhelpful behaviors and experiences related to emotional overcontrol; including psychological inflexibility, suppression of emotions, and experience of loneliness.

The participants will undergo the treatment with RO-DBT, which include an approximately 22 week long treatment consisting of individual psychotherapy and parallel skills training in group.

BACKGROUND

Eating disorders are psychiatric disorders characterized by loss of control over food intake. Eating disorders impair function, have serious health implications, and are associated with high mortality. The mortality in anorexia nervosa is especially striking, as it is one of the psychiatric disorders with the highest mortality rates. Approximately 20% of patients with eating disorders develop long lasting illness, referred to as Severe and Enduring Eating Disorders (SEED). However, research on this group is scarce, and knowledge about treatment options is extremely limited.

Clinical factors related to poor outcome in anorexia, including SEED, are low age at onset, premorbid obsessive-compulsive personality traits, and autistic traits. Further, dysfunctional emotion regulation (ER) has been identified as a transdiagnostic psychological risk factor for many psychiatric disorders, including eating disorders. Dysfunctional ER can be characterized by traits of excessive self-control and emotional inhibition (overcontrol). Some propose that dysfunctional ER, characterized by emotional overcontrol, is a core aspect of restrictive eating disorders such as anorexia. Further, research suggests that loneliness relates to ER, with ER strategies like excessive self-control and emotional overcontrol in youths being associated to an increased sense of social isolation and lower life satisfaction later in life, increasing the risk for enduring mental health problems.

Despite this, few studies have evaluated the feasibility and effect of transdiagnostic interventions that specifically address emotional overcontrol in clinical trials, with the specific aim to evaluate if this might have the potential to change a chronic course. Radically open dialectical behavior therapy (RO-DBT) is a treatment which has been developed to target emotional overcontrol, hypothesized to be an underlying cause of the development and maintenance of anorexia and other restrictive eating disorders. In previous studies, the research group has found more pronounced emotional overcontrol in patients with anorexia compared to bulimia, that adult patients with anorexia responded to psychological treatment with RO-DBT targeting overcontrol, and that the patients appreciated this focus in therapy. Further, one previous study has evaluated RO-DBT for adolescents with eating disorders in an uncontrolled study with positive results. However, more research is needed.

The aim of this study is to evaluate feasibility and effect of the psychological treatment RO-DBT for adolescents with primarily restrictive eating disorder symptoms that have not responded to previous treatments and who therefore are at risk for developing SEED.

Primary hypotheses

1. The RO-DBT treatment is feasible in an outpatient, adolescent, clinical setting.
2. After treatment with RO-DBT the patients will reduce their eating disorder symptoms, and, if underweight, regain weight.

Secondary hypotheses

1. Patients self-perceived health and positive feelings in social situations will increase during treatment.
2. Patients functional impairment, maladaptive overcontrol and experienced level of loneliness will decrease during treatment.
3. Patients comorbid symptoms of anxiety and depression will decrease during treatment.
4. Changes in eating disorder symptoms and maladaptive overcontrolled behaviors will be an effect of the intervention, i.e. they will be observable after introduction of the intervention.
5. The effects will remain at one-year follow-up.

PROCEDURE

Patients between 14 and 19 years old will be recruited from the Uppsala Child and adolescent psychiatry or the Adult psychiatry, either from the Eating disorder unit or the General psychiatry unit.

Anorexia is overrepresented among eating disorder patients with maladaptive overcontrol. However, as eating disorder symptoms often change in symptom expression over time, potential participants are not limited to those with a diagnosis of anorexia during the baseline assessment. Therefore, adolescent patients with eating disorder symptoms that have not responded fully to eating disorder treatment or who have relapsed after recovery will be provided with information about the treatment study, either by their ordinary caregiver or through advertisement in waiting rooms. Those interested can report this to their caregiver who contacts the research group, or patients can contact the research group themselves. Thereafter, the family is contacted via phone and booked for an information and evaluation sessions where additional verbal information is provided and the information to research persons is handed to the youths and their caregivers. If the participant is still interested and identified as overcontrolled, assessed with the The Overcontrolled Global Prototype Rating Scale developed by Lynch (2018), fulfill all inclusion but not exclusion criteria, he/she is invited to participate in the study. Informed consent is gathered from bort the participant and his/her caregivers. Participants should have gone through the diagnostic procedure at the Eating disorder unit during the last three months before enrollment in the study, if not, the same diagnostic procedure will be part of the study.

After treatment, participants will be interviewed regarding their experiences of the treatment.

STUDY DESIGN

The study is a single-case experimental design study with multiple baselines, where each individual serves as their own control. Participants will be randomized to baseline length. Baseline measures on 4-6 weeks before treatment will be assessed and contrasted to measures weekly during treatment. Follow-up will be conducted one year after treatment termination.

QUALITY ASSURANCE PLAN

Before the intervention, all therapists will be trained by authorized trainers in RO-DBT and will have provided RO-DBT to at least one individual patient. During the intervention, the team will be supervised by an expert-led consultation team. Adherence to treatment protocol will be assessed via therapist self-reports of session content which will be contrasted to the treatment outline recommended by Lynch (2018).

SAMPLE SIZE

In single-case studies, at least three replications are recommended to achieve generalizability. An analysis is conducted with three patients, followed by replication in three new patients, and so on. This means 3x3=9, which is the minimum number of participants recommended. Since dropouts are common in this patient group, 18 patients will be included.

STATISTICAL AND QUALITATIVE ANALYSES

Feasibility will be analyzed by presenting descriptive data on implementation success, treatment credibility/expectancy and adverse events. Treatment acceptability and satisfaction will be assessed by interviewing participants, and the transcripts of these interviews will be analyzed using thematic analysis..

Quantitative, repeated, data will be analyzed visually and statistically according to recommendations for single-case research. Symptoms measured weekly during a baseline before treatment (phase A) will be contrasted to weekly ratings during treatment (phase B) for each participant. For the visual analyses, the steps outlined by Kratochwill and colleagues (2013) will be followed, which include examining the level, trend, variability, immediacy of effect, overlap, and consistency of data patterns. For statistical analyses, Tau-U will be utilized as a measure of effect size. Multiple baselines between individuals will be analyzed to control for time-effects.

Within-group analyses will be calculated for the measures before treatment, after treatment, and at one-year follow-up. Reliable change index will be calculated to assess whether the changes are statistically significant and meaningful.

Inclusion Criteria:

• Being 14-19 years old
• Primarily restrictive eating disorder symptoms that remain despite at least one previous treatment attempt.
• Problems related to maladaptive overcontrol
• Written informed consent (for minors, this includes consent from all caregivers and the minors themselves).

Exclusion criteria:

• Eating disorders symptoms in need of emergency care
• High risk for suicide
• An inability to respond to the questionnaires or participate in the skills class, e.g., due to lack of knowledge in Swedish.

Data will not be made publicly available due to confidentiality, but can be made available upon reasonable request to the corresponding author.

Blinded randomization of baseline length is performed in blocks of three (4, 5, or 6 weeks) by a researcher or research assistant not otherwise involved with the participants.