Predictors of dropout and remission in family therapy for adolescent anorexia nervosa in a randomized clinical trial




The purpose of this study is to explore the predictors of dropout and remission in the treatment of adolescent anorexia nervosa (AN) using family therapy.


Data derived from a randomized clinical trial comparing short and long term family therapy for adolescents with AN were used. A rotated component analysis was employed to reduce the number of variables and to address problems of collinearity and multiple testing. Dropout was defined as participating in less than 80% of the assigned therapy. Participants were classified as remitted if they obtained an ideal body weight greater than 95% and a global eating disorder Examination score within two standard deviations of community norms at the end of 12 months.


Co-morbid psychiatric disorder and being randomized to longer treatment predicted greater dropout. The presence of co-morbid psychiatric disorder, being older, and problematic family behaviors led to lower rates of remission. A reduction of child behavioral symptoms, a decline in problematic family behaviors, and early weight gain were all within treatment changes that increased the chance of remission.


Co-morbid psychiatric disorder, family behaviors, and early response to treatment are important factors when predicting dropout and remission in family therapy for adolescent AN. © 2006 by Wiley Periodicals, Inc. Int J Eat Disord 2006; 39:639–647