Stability of eating disorder diagnoses: A commentary

Abstract

Eat ing disorders, including anorexia nervosa (AN), bulimia nervosa (BN), and binge eating disorder (BED), are one of the most prevalent psychiatric problems for girls and young women, result in subjective distress and functional impairment, and increase risk for future onset of obesity, depression, suicide attempts, anxiety disorders, substance abuse, and health problems. 1,2 Although these eating disorders are clearly associated with functional impairment, less is known about the course of these eating disorder diagnoses over time. The purpose of this commentary is to discuss the findings of two articles on the longitudinal course of eating disorder diagnoses that appear in this special issue to discuss implications for nosology, and to offer suggestions regarding future research in this area.Long-term Stability of Eating Disorder Diagnoses (Fichter, & Norbert) This manuscript reports data on the stability of eating disorders over a 12-year period in a large treatment sample (N 5 311). Assessments were conducted at the beginning of therapy and at 2-year, 6-year, and 12-year follow-ups. The 12-year stability rate was 41.4% for AN, 34.4% for BN, 3.0% for BED, and 3.2% for other eating disorder not otherwise specified diagnoses (EDNOS). Results suggest moderate longitudinal instability of AN and BN, and pronounced longitudinal intability of BED and other EDNOS conditions. This instability was due to recovery, relapse, and diagnostic crossover. Most crossovers occurred from AN to BN and from BED to BN. Long-term outcome for BN and BED was encouraging, with both showing high recovery rates. The long-term outcome for AN was worse relative to BN and BED. Furthermore, the longitudinal course of BN and BED seemed to be nearly identical with similar percentages of remission and relapse. Relapse rates were low in this treatment sample.Before discussing the implications of these findings for nosology and future research directions, it is important to compare the findings from this study with other studies that have been previously published. Although research indicates that the overarching diagnosis of ''eating disorder'' is relatively stable, 3 several studies suggest that there is considerable longitudinal instability of the specific eating disorder diagnoses of AN, BN, and EDNOS in treatment samples, resulting from recovery, relapse, and crossover between and within the eating disorder diagnoses. 3,4-6 The recovery rates for AN and BN were moderate to high at each of the follow-up assessments in Fichter and Quadflieg study, which is similar to the rates reported by 7 who found that at 5-year follow-up, 56% of patients with AN, 74% of patients with BN, and 78% of patients with EDNOS no longer met criteria for any eating disorder diagnosis.Recovery rates of other treatment studies, however, are somewhat lower. For example, Fichter and Quadflieg reported a remission rate of 43.8% in AN and 56.8% in BN over a 2-year period; these recovery rates increased to about 60% for AN and about 70% for BN over a 6-12-year period....

Bulimia nervosa with history of anorexia nervosa: Could the clinical subtype of anorexia have implications for clinical status and treatment response?

“…In addition, they had more severe bulimic symptoms at the beginning of the treatment (higher initial scores in the BITE and the PSRs) and a greater tendency to present a lack of response after 6 months of treatment (29% of BN/AN1 patients vs. 6% of BN/AN2 did not respond). Although some authors have pointed out that it is not the diagnosis but low body weight or longer duration of more severe eating disorder which are related to worse outcome, 34,35 our findings support the idea that a history of AN in BN patients can be a factor associated to a protracted illness. In conclusion, our Study #1 replicates the data reported in literature, which defend the existence of differences between BN patients with/without a history of AN.…”

“…It is also possible that impulsivity, which was higher in our BN/AN1 R patients than in our BN/AN1 BP patients, might be related to the poorer outcome observed in the former, as it has been widely demonstrated that impulsivity is a strong predictor of bad outcome in BN 12,34,[43][44][45][46][47][48] In conclusion, it has been suggested that studies should be designed to determine whether the diagnosis of AN or certain aspects of AN may be capable of determining clinical characteristics or outcome in patients with BN. 35 Our study should be considered a modest contribution to this line of thinking. We have considered the subtype of previous AN as a differential criterion which could be of interest in order to subtype patients with BN.…”

Long‐term outcome of pediatric eating pathology and predictors for the onset of loss of control over eating following weight‐loss treatment

“…A second strength is that we investigated ED symptoms instead of ED diagnoses, as it was already recommended by Stice and Spoor. 43 Next, the use of structured clinical interviews to assess eating-disordered behaviors was another important asset of this study. A final strength was the inclusion of a sample that consisted of both boys and girls.…”