Testing Two Treatment Approaches and the Therapeutic Alliance

New information emerges about the causal direction of treatment for patients with bulimia nervosa.

The importance of a strong therapeutic alliance for good outcome in treatment of bulimia nervosa (BN) was recently underscored in results from a two-site study of 80 adults (J Consult Clin Psychol. 2015. 83:637). The study examined whether therapeutic alliance affected the outcome in two psychosomatic treatments.

The adults, all of whom had DSM-5 diagnoses of BN or partial BN (subjective rather than objective binge-eating episodes), were randomized to integrative cognitive-affective therapy (ICAT) or enhanced cognitive-behavioral therapy (CBT-E). ICAT, a recently developed treatment approach, emphasizes the role of emotion in bulimic behavior and also argues that bulimic behavior regulates emotion. Thus, interventions are aimed at identifying cues for binge eating and finding ways to help patients manage urges to binge. ICAT aims to identify cues for binge eating patterns and attempts to modify behaviors that lead to negative emotions. CBT-E is a revised version of DBT for BN that uses psychoeducation, self-monitoring, and behavioral exposure to normalize eating patterns and to attempt to modify cognition, especially overvaluation of shape and weight. In this study both groups had 50-minute treatment sessions over 17 weeks.

At each session and at the end of treatment, bulimic symptoms, such as frequency of binge eating and purging, were assessed. Therapeutic alliance was measured at sessions 2, 8, and 14, and 1 week from the end of treatment.

The importance of the patient’s own characteristics

Most of the 80 participants were white (88%) and female (90%); mean age was 27 years, and mean body mass index (BMI) was 23.9 kg/m².The authors found that patients reported strong therapeutic alliances that were stable over time; however, stronger overall therapeutic alliances led to greater reduction in BN symptoms. In turn, greater reductions in BN symptoms predicted improved therapeutic alliance. For example, greater depression, anxiety, and emotional dysregulation were associated with lower overall alliance. There were interactions between coexisting psychological variables and alliance. Participants with fewer symptoms reported stronger overall therapeutic alliances in the ICAT group, while those with greater symptomatology reported stronger overall therapeutic alliances in CBT-E.

The results suggest better therapeutic alliance leads to reduced BN symptoms, which then leads to improvements in alliance. The authors do note that the small size of the study (80 patients and 4 therapists) limited the study’s ability to address the importance of the patient‘s and therapist’s individual contributions to the alliance. Nonetheless, the findings underscore the value of a strong therapeutic alliance.