(PS6-A24) Investigating the Utility of Pictorial Figure Rating Scales for Identifying Eating Disorder Risk Among College-age Females

Background: Existing eating disorder (ED) screeners focus on evaluating behavioral symptoms of EDs; however, measuring cognitive risk factors that may emerge before behavioral symptoms, such as body dissatisfaction, may allow for the early identification of high-risk individuals. Pictorial figure rating scales can provide a holistic measure of body dissatisfaction across body shape ideals and have the potential to serve as a non-burdensome screening tool for healthcare providers assessing ED risk. Therefore, the current study investigated pictorial scales of body dissatisfaction (thin, curvy, muscular) as a way to screen for ED risk in female college students. It was anticipated that greater discrepancies across all body shapes would accurately identify female college students at high risk for an ED.

Methods: The current study is a secondary data analysis of a sample of 264 female college students (18-25 years) who self-identified as White (38.2%), Black (32.6%), or Hispanic (29.2%). The Binge Eating, Purging, and Restriction subscales of the Eating Pathology Symptom Inventory (EPSI) measured ED symptoms. Body dissatisfaction was measured using actual-ideal discrepancy scores for thinness (Stunkard Figure Rating Scale), curviness (Curvy Ideal Silhouette Scale: Curvy matrix), and muscularity (Body Image Matrix of Thinness and Muscularity: Muscularity matrix).

Data Analytic Plan: EPSI subscale scores greater than or equal to the mean reported in an outpatient sample of ED patients were coded as at risk. The area under the curve (AUC) was estimated and compared for each receiver operating characteristic curve analysis using a 95% confidence interval to determine whether pictorial scales would be good predictors of clinically significant levels of binge eating, purging, and restricting behaviors.

Results: Thinness body discrepancy scores demonstrated low-moderate accuracy for predicting ED risk status for binge eating (AUC =.66, 95% CI: .60-.73), purging (AUC =.71, 95% CI: .64-.78), and restriction (AUC =.44, 95% CI: .37-.51). Muscularity body discrepancy scores demonstrated low accuracy for predicting ED risk status for binge-eating (AUC =.53, 95% CI: .46-.60), purging (AUC =.58, 95% CI: .49-.66), and restriction (AUC =.48, 95% CI: .41-.55). Curvy body discrepancy scores demonstrated low accuracy in predicting ED risk status for binge-eating (AUC =.50, 95% CI: .43-.58), purging (AUC =.50, 95% CI: .42-.59), and restriction (AUC =.50, 95% CI: .43-.58). Exploratory analyses revealed that accuracy in ED risk identification across all body shape discrepancies did not significantly differ by racial identity (ps >.05).

Discussion: Findings suggest that body dissatisfaction, as measured via body shape discrepancy, was a poor indicator of ED risk among female college students. Given that body dissatisfaction is a strong risk factor for the development of EDs, future research should examine whether body dissatisfaction remains a poor predictor of ED risk using alternative assessment methods.