Understanding eating disorders within internalizing psychopathology: A novel transdiagnostic, hierarchical-dimensional model

2017 
Abstract Background Several problems with the classification and diagnosis of eating disorders (EDs) have been identified, including proliferation of ‘other specified’ diagnoses, within-disorder heterogeneity, and frequent diagnostic migration over time. Beyond problems within EDs, past research suggested that EDs fit better in a spectrum of internalizing psychopathology (characterized by mood and anxiety disorders) than in a separate diagnostic class. Purpose To develop a transdiagnostic, hierarchical-dimensional model relevant to ED psychopathology that: 1) reduces diagnostic heterogeneity, 2) includes important dimensions of internalizing psychopathology that are often excluded from ED diagnostic models, and 3) predicts clinical impairment. Procedures Goldberg's (2006) method and exploratory structural equation modeling were used to identify a hierarchical model of internalizing in community-recruited adults with EDs ( N  = 207). Findings The lowest level of the hierarchy was characterized by 15 factors that defined specific aspects of eating, mood, and anxiety disorders. At the two-factor level, Internalizing bifurcated into Distress (low well-being, body dissatisfaction, suicidality, dysphoria, ill temper, traumatic intrusions) and Fear-Avoidance (claustrophobia, social avoidance, panic symptoms, dietary restricting, excessive exercise, and compulsions). Results showed that the lowest level of the hierarchy predicted 67.7% of the variance in clinical impairment. In contrast, DSM eating, mood, and anxiety disorders combined predicted 10.6% of the variance in impairment secondary to an ED. Conclusions The current classification model represents an improvement over traditional nosologies for predicting clinically relevant outcomes for EDs.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    75
    References
    34
    Citations
    NaN
    KQI
    []