Eating-specific interpersonal difficulties: changes and impacts on outcomes during ten-session cognitive behavioral therapy for eating disorders (CBT-T)

Objective

The relationship between self-reported interpersonal difficulties and eating disorder symptoms is well-established. The Interpersonal Relationships in Eating Disorders (IR-ED) is a new measure of eating-specific interpersonal difficulties (food-related isolation, avoidance of body evaluation, foot-related interpersonal tension). This study aims to (1) explore changes in IR-ED scores, eating disorder symptoms, and psychosocial impairment during ten-session cognitive behavioral therapy for eating disorders (CBT-T), (2) assess whether pre-treatment IR-ED scores moderate the trajectory of change in eating disorder symptoms and psychosocial impairment during CBT-T and (3) investigate the relationship between changes in the IR-ED and changes in eating disorder symptoms and psychosocial impairment.

Method

Individuals with eating disorders (N = 126, Mage = 26.08) received CBT-T and completed questionnaires at pre-, mid-, and post-treatment.

Results

Eating-specific interpersonal difficulties, eating disorder symptoms, and psychosocial impairment reduced during CBT-T. Severity of pre-treatment eating-specific interpersonal difficulties was unrelated to change in eating disorder symptoms or psychosocial impairment, irrespective of diagnosis or body mass index. Changes in interpersonal difficulties were associated with concurrent changes in eating disorder symptoms and psychosocial impairment. Early change in interpersonal difficulties did not predict later change in eating disorder symptoms or psychosocial impairment, and early changes in symptoms or psychosocial impairment did not predict later changes in interpersonal difficulties.

Conclusions

Eating-specific interpersonal difficulties improve during CBT-T, and individuals benefit from treatment regardless of their pre-treatment interpersonal difficulties. Theoretical and clinical implications are discussed.