(PS11-83) Model Cross Validation: body Dysmorphic Disorder Within the Hitop Taxonomy

Body Dysmorphic Disorder  s both under identified and associated with high rates of suicide which make improved diagnosis and treatment critical issues. BDD is an enigma, however, sharing compulsions with the obsessive compulsive and related disorders (OCD-RD) and sharing body-image disturbances with the eating disorders. HiTOP, a transdiagnostic and dimensional approach to psychopathology, can inform both clinical and scientific issues by allowing us to examine the validity of BDD’s assignment to the OCD -RD spectrum rather than to the eating disorders (ED) spectrum. Our study is the first to address this empirically with a HiTOP conceptualization. Accordingly, we evaluated a bifactor model of lower order internalizing disorder factors that all shared a general factor of psychopathology. This model was cross validated with two samples to test the hypothesis that the symptom marker of BDD was more related to the ED spectrum, rather the OCD-RD spectrum.

Method

Sample and Procedure

364 undergraduates from Midwest colleges and universities participated in an online survey.

Measures

Spectra             Disorder/Symptom Markers

Sadness            GAD:GAD-7; Depression: CES-D 10; PTSD:PCL-C

Fear:                 Agoraphobia: SMAA; Panic Disorder: SMPD; Social Anxiety: SOPHS

OCD-RD.         OCD: OCI-R; Trichotillomania: MGH; Excoriation: MSPS; Hoarding: SI-R

ED:                  Eating Disorder: BSQ-8C; Body Dysmorphic: DCQ; Binge Eating: BEDS-7

Analysis and Results

In our Confirmatory Factor Analysis of a bi-factor model  all items loaded onto a single factor of General Psychopathology as well as their respective domain of either Sadness Disorders, Fear Disorders, Body Image & Eating Disorders, or OCD – Related Disorders. A nested chi square difference (chi sq diff) test demonstrated that Body Dysmorphia fit better as an indicator of Body Image & Eating Disorders (chi sq diff (1) = .012, p = .913) than OCD-Related Disorders (chi sq diff (2) = 104.02, p < .001). We cross-validated this structure on random halves of the sample through invariance testing, which passed at the weak and strong level.

Discussion

The bifactor CFA model with BDD loading on the ED factor was the best fit to the data in parent and cross validating child models. These findings suggest that BDD’s placement in the hierarchy may warrant greater attention.