Category: Eating Disorders

PS1- #C88 - Shared Vulnerabilities for Social Anxiety, Binge Eating, and Obesity

Friday, Nov 17
8:30 AM – 9:30 AM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Eating Disorders | Social Anxiety | Obesity / Overweight

Social anxiety, binge eating, and obesity are highly comorbid (Darlymple et al., 2015), but little is known about mechanisms of their overlap. The research on shared vulnerability between social anxiety and disordered eating identifies that maladaptive perfectionism (MP), social appearance anxiety (SAA), and fear of negative evaluation (FNE) are shared vulnerabilities (Levinson & Rodebaugh, 2016). Negative affect (NA) and dietary restraint relate to both binge eating and obesity.  (Mason et al., 2016). The literature on vulnerabilities shared between obesity and social anxiety/binge eating is lacking. Although most of these risk factors are associated with social anxiety, binge eating, and obesity independently, they have not been tested in a combined model.  

In the current study, women (N=300) completed measures of social anxiety, binge eating, FNE, SAA, restraint, NA, MP, and had their body mass index (BMI) assessed at two time points, six months apart. Utilizing structural equation modeling we tested five potential vulnerabilities (FNE, SAA, restraint, NA, and MP) as predictors of three outcomes: social anxiety, binge eating, and obesity (i.e., BMI) risk.  We tested both a (a) cross-sectional model and (b) prospective 6-month model.

Fit for the cross-sectional model was excellent (CFI=1.0, TLI=1.02, RMSEA < 0.001, SRMR=.01). All five predictors (FNE, SAA, restraint, NA, and MP) were associated with social anxiety; SAA, restraint, and NA were significantly associated with binge eating; restraint and NA were significantly associated with BMI (all ps < .05). Fit for the prospective model was also excellent (CFI=1.0, TLI=1.01, RMSEA < 0.001, SRMR=.03). MP predicted social anxiety (p=0.008) and SAA predicted binge eating over time (p=0.003). There were no prospective predictors of BMI.

At one time point we found that restraint and NA were associated with all three outcomes. However, in a prospective model we did not find evidence of shared vulnerabilities across social anxiety, binge eating, and obesity. MP predicted later social anxiety, but not binge eating. These results are consistent with previous research showing that MP prospectively predicts social anxiety (Levinson & Rodebaugh, 2016). Our analysis further supported SAA, but not NA or restraint, as a predictor of binge eating (Brosof & Levinson, 2016). No vulnerability predicted obesity across time. More research is needed to identify what vulnerabilities are shared between obesity, social anxiety, and binge eating. It is possible that traits not assessed here, such as self-esteem and weight stigma, may be shared across disorders. Understanding mechanisms of overlap may improve prevention and treatment efforts by focusing interventions on shared vulnerabilities, thereby treating multiple problems simultaneously.  

Irina Vanzhula

Graduate student
University of Louisville

Cheri Levinson

University of Louisville