The role of readiness and confidence to change in the treatment of atypical anorexia nervosa and bulimia nervosa

Objective: Atypical anorexia nervosa (AN) is a serious eating disorder that is more common in the population than AN. Despite this, people with atypical AN are less likely to be referred or admitted for eating disorder treatment and there is evidence that they are less likely to complete or benefit from existing interventions. This study examined whether baseline readiness and/or confidence moderated outcomes from 10-session cognitive behavioral therapy among people with atypical AN and bulimia nervosa (BN), and whether the impact of these variables differed between diagnoses. 

Methods: Participants (n = 67; 33 with atypical AN) were a subset from an outpatient treatment study. Linear mixed model analyses were conducted to examine whether baseline readiness and/or confidence moderated outcomes. 

Results: People with BN who had higher levels of readiness or confidence at baseline had steeper decreases in eating disorder psychopathology over time. There was no evidence that readiness or confidence moderated outcomes for people with atypical AN. 

Discussion: This study suggests that the moderators that have been identified for other eating disorders may not apply for people with atypical AN and highlights a need for future work to routinely investigate whether theoretically or empirically driven variables moderate outcomes in this little-understood population. 

Public Significance Statement: People with bulimia nervosa with higher readiness and confidence experienced greater decreases in eating disorder symptoms than people with lower readiness and confidence when treated with cognitive behavioral therapy. These findings did not apply to people with atypical anorexia nervosa. Results demonstrate that future work is urgently required to identify helpful treatments for people with atypical anorexia nervosa as well as the variables that have a positive impact on outcomes in treatment for these individuals.