Background. One hundred and thirty-five women with bulimia nervosa participated in a randomized clinical trial designed to determine whether the addition of exposure with response prevention to a core of cognitive-behavioural therapy (CBT) leads to greater clinical improvement and lower risk of relapse. We present results from the end of treatment and 6- and 12-month follow-up.

Methods. Participants received eight sessions of CBT and were then randomized to either exposure to pre-binge cues (B-ERP), exposure to pre-purge cues (P-ERP), or a relaxation training control condition (RELAX).

Results. CBT produced significant clinical change. At the end of the behavioural treatments, there were no significant differences across the three groups on abstinence (66% in B-ERP, 45% in P-ERP and 47% in RELAX), or frequency of bingeing and purging. B-ERP, but not P-ERP, significantly reduced anxiety on the cue reactivity assessment, food restriction, body dissatisfaction and depression. These differences were not maintained at 6-month follow-up. At 12-months, B-ERP was independently associated with lower food restriction and better global functioning.

Conclusions. CBT is a highly effective treatment for bulimia nervosa. B-ERP was modestly superior to P-ERP at post-treatment; however, the advantage did not remain throughout the follow-up interval. ERP for bulimia nervosa is an expensive and logistically complicated treatment that does not appear to offer any significant additive benefits that are proportional to the amount of effort required to implement the treatment.