Group psychotherapy for eating disorders: A randomized clinical trial and a pre-treatment moderator and mediator analyses

English summary

The aim of this thesis was twofold. First, I wanted to examine the effect of client feedback on

treatment attendance and outcome in group psychotherapy for eating disorders. Second, I wanted

to contribute to the relatively scarce body of research on the consequences of an eating disorder

on functionality.

Group psychotherapy of various types is widely used to treat individuals with eating

disorders. Dropout is, however, an important problem in the treatment of these patients, and it is

therefore important to find ways to increase attendance. One of the means to address nonattendance

and dropout in psychotherapy is by monitoring patient progress, and using the

information to make treatment alterations in collaboration with the patient. As described in Papers I

and III, we planned to assess the effect of client feedback on attendance and outcome for patients

with eating disorders in group therapy. We conducted a randomized clinical trial and included 159

adult participants, 156 females and 3 males, diagnosed with bulimia nervosa, binge eating

disorder, or eating disorder not otherwise specified according to DSM-IV. Eighty participants were

allocated to the experimental group, and 79 participants were allocated to the control group. In

both groups, the participants received 20-25 weekly group psychotherapy sessions. In the

experimental group, participants gave and received feedback about the therapy progress and

alliance, measured before and after each session using the Outcome Rating Scale and the Group

Session Rating Scale. The primary outcome was rate of attendance at treatment sessions; the

secondary outcome was the severity of eating disorder symptoms measured with the Eating

Disorder Examination interview. Exploratory outcomes were psychological distress measured with

the Symptom Checklist-90-R, social functioning measured with the Sheehan Disability Scale, and

episodes of self-harm and suicide attempts measured with a modified version of the Self-Harm

Inventory. Results showed that feedback had neither a significant effect on the rate of attendance

(p = .96), nor did it improve outcome (all p values > .05). The results point to the importance of

organizational flexibility, and to the therapists’ view of the usefulness of the feedback measures.

One appropriate approach to a better understanding of functional impairment

associated with eating disorders is to disentangle the complex interplay of specific factors

contributing to the extent of functional impairment. Thus, while the link between eating disorder

severity and functional impairment has been established in other studies, it is likely that this

relationship is moderated and mediated by a number of other pre-treatment characteristics

associated with eating disorders. In paper II, I therefore examined duration of illness and body

mass index as possible moderators of the relationship between eating disorder severity and

9

functional impairment, as well as psychological distress as a possible mediator of this relationship.

The same 159 patients included in the F-EAT trial were included in this study. Regression analysis

was applied to assess the effect of the hypothesized moderators and mediators. Eating disorder

severity was measured with the Eating Disorder Examination Questionnaire, functional impairment

was measured with the Sheehan Disability Scale, and psychological distress was measured with

the Symptom Checklist-90-R. Duration of illness and body mass index was assessed clinically.

Results showed that duration of illness significantly moderated the relationship between eating

disorder severity and functional impairment, and the relationship between psychological distress

and functional impairment. In both cases, the relationship was strongest for patients with a shorter

duration of illness. Psychological distress partly mediated the relationship between eating disorder

severity and functional impairment. Body mass index was not a significant moderator of the

relationship between eating disorder severity and functional impairment. Overall, this study

established a link between eating disorder severity, psychological distress and functional

impairment indicating that both eating disorder severity and psychological distress are more

strongly related to impaired role functioning for patients with more recent onset of an eating

disorder.