Social anxiety and eating disorder comorbidity and underlying vulnerabilities: Using network analysis to conceptualize comorbidity

OBJECTIVE: Eating disorders (EDs) and social anxiety disorder (SAD) are highly co-occurring. This comorbidity is extremely relevant, given that individuals with comorbid ED-SAD are less likely to seek and/or benefit from ED treatment. METHOD: We used network analysis to conceptualize ED-SAD comorbidity in a sample of 2,215 participants with a primary diagnosis of ED, SAD, or no known diagnosis. We used novel network analyses methods to select symptoms for our models, identify potential illness pathways (i.e., bridge symptoms) between disorders and underlying vulnerabilities (e.g., perfectionism, social appearance anxiety), and to compare across sample types (e.g., clinical vs. nonclinical). We also tested several novel network analyses methods aimed at the following methodological concerns: (a) topological concerns (i.e., which items should be included in NA models), (b) how to use empirical indices to quantify bridge symptoms and (c) what differences in networks across samples mean. RESULTS: We found that difficulty with drinking beverages and eating in public were bridge symptoms between ED and SAD. We also found that feeling nervous about one's appearance was a bridge symptom. CONCLUSIONS: We identified public eating and drinking as bridge symptoms between EDs and SAD. Future research is needed to test if interventions focused on public eating and drinking might decrease symptoms of both EDs and SAD. Researchers can use this study (code provided) as an exemplar for how to use network analysis, as well as to use network analysis to conceptualize ED comorbidity and compare network structure and density across samples.