The definition and assessment of remission in anorexia nervosa (AN) needs greater consensus. Particularly in adolescents, the use of patient-reported composite indices (such as the Eating Disorder Examination [EDE] Global Score) as the sole measure of psychological remission has the potential to obscure patients' true clinical status, given developmental factors and the propensity towards symptom minimization in AN. End of treatment (EOT) data from a randomized controlled trial comparing two formats of manualized family-based treatment for adolescents with AN (N = 106) were analyzed. Participants completed the EDE, and their parents completed a parent-as-informant version of the EDE (Parent Eating Disorder Examination; PEDE). Rates of remission were compared across indices (i.e., EDE Global Score vs. diagnostic item analysis) and informant (i.e., adolescent vs. parent), both independently and in combination with the achievement of a percent median body mass index (% mBMI) greater than or equal to 95%. For both adolescent and parent reports, there were higher rates of remission when defined by Global Score than when defined by EDE or PEDE diagnostic items. There were no significant differences in remission rates based on informant. In the assessment of remission in AN, the EDE Global Score may not detect some adolescents who continue to exhibit clinically significant psychological symptoms. This study supports a detailed, multidimensional approach to assessing remission in adolescent AN to optimize sensitivity to patients' diagnostic profile. Future research should explore whether parent-child concordance on measures of ED psychopathology varies over the course of treatment.