Predicting Treatment Outcome for Teens with Bulimia Nervosa

Many types of effective treatment for bulimia nervosa (BN) have been developed and tested among adults.  However, this is not the case for teens with BN, even though many of the behaviors that mark BN first occur in adolescence, according to Daniel Le Grange, PhD and colleagues (J Am Child Adolesc Psychiatry. 2008; 47:464).

Dr. Le Grange and co-workers designed a randomized controlled trial of 80 teens to compare factors that predict success or failure of family-based treatment and  individual supportive psychotherapy for BN. Treatment was provided on an outpatient basis and consisted of 20 sessions, 50 minutes each, over 6 months. Sessions were conducted weekly during the first 3 months of therapy, every second week for the next 2 months, and every 3 weeks for the remainder of the study.

Two male teen patients (2.5%) and 78 female adolescents (97.5%) with a mean age of 16.1 years and a mean body mass index of 22.1 kg/m2, participated in the study. Forty-six of the participants were from intact two-parent families. Thirty-seven of the participants met DSM-IV criteria for BN, and the remainder met strict criteria for partial BN. (Partial BN was defined as having some, but not all, of the criteria for a diagnosis of BN.)

Two types of treatment 

Family-based treatment focused on motivating parents to help their adolescents overcome the eating disorder. This manual-based treatment proceeds through three phases. The first phase is designed to help parents, in collaboration with their children, to disrupt binge eating, purging, restrictive dieting, or any other pathologic methods the teens may be using to control their weight. The second phase begins once the teen is abstaining from disordered eating. In this stage, the parents are encouraged to turn their influence over eating behaviors back to the teen. In the third phase, attention is placed on ways in which the family can help the teen deal with the impact that the eating disorder has had on the developmental processes.

Supportive psychotherapy explores underlying problems and is manual-based, goal oriented, and the patient is encouraged to take the lead to talk about issues of concern.

What predicted outcome after treatment?

The authors found several factors that predicted the success of treatment.  First, lower levels of concern about eating are the best predictor of remission for teens with BN.  Next, the use of family-based treatment of BN may be most effective for teens with low levels of eating disorder psychopathology. Four variables on the Eating Disorders Examination had a significantly moderatingsignificantly moderated the effects of treatment on partial remission: weight concerns, shape concerns, eating concerns, and EDE global score. In each case, teens receiving individual supportive psychotherapy showed similar rates of partial remission regardless of their EDE scores. In all cases, participants receiving supportive psychotherapy had similar rates of partial remission, regardless of scores on the EDE. In contrast, for teens receiving family-based therapy, partial remission rates were much higher for those with low EDE scores.

Stressing that their study was exploratory only, the authors note that the clearest predictor to emerge from the study was that participants with lower eating concerns scores at baseline were more likely to have remitted (abstained from binge eating and purging) after treatment and at follow-up, regardless of which type of therapy they had. They also noted that some factors that have traditionally been viewed as predictive of treatment outcome, such as duration of illness, age, and diagnosis, were neither predictors nor moderators of outcome in this study.