5 - (IOP 35) A Network Analysis of Avoidant/restrictive Eating, GI Symptoms, and the Impact of a Gluten Free Diet in a Celiac Disease Sample

Background: Individuals with celiac disease (CD) are at risk for eating disorders (ED) characterized by weight and shape concerns (i.e., weight/shape disordered eating) and food avoidance driven by aversive emotional, physiological reactions to food itself or the acting of eating (avoidant/restrictive eating). Mechanisms behind this comorbidity are not well understood. Network analysis is a statistical technique utilized in psychopathology research to examine interrelationships among symptoms. Network analysis allows for the identification of symptoms that are most central to a disorder. Central symptoms are measured by strength centrality and are theorized to activate other symptoms through their strong connections. Bridge symptoms are those that are highly connected to symptoms within others cluster and reflect pathways linking different disorders. The aim of the present study is to use undirected network analyses to examine the psychopathology of gastrointestinal (GI) symptoms, ED symptom severity, and the psychosocial impact of maintaining a gluten free diet in individuals with CD.

Methods: The sample consist of 397 patients with CD (n = 397; mean age (SD) = 41.05 (17.05) years) who presented to a GI clinic. Symptoms were assessed using the Gastrointestinal Symptom Rating Scale (GSRS), the Impact of a Gluten Free Diet Questionnaire (IGFDQ), the Nine Item ARFID Screen (NIAS), the Eating Disorder Examination Questionnaire 7 (EDE-Q7), the Generalized Anxiety Disorder 2-item (GAD-2), and the Patient Health Questionnaire-2 (PHQ-2). We constructed a cross-sectional graphical LASSO network.

Results: The strength CS coefficient (1,000 bootstraps) was good for the network (strength [S] = 0.75). Bridge expected influence (BEI) also had good stability for the network (BEI stability = 0.75). Taken together, the results of the network analysis can be reliably interpreted. The symptoms with the greatest strength centrality were pain (S = 1.27), overvaluation of weight and shape (S = 1.21), less freedom (S = 1.08), appetite (S = 0.83), bloating (S = 0.78), body dissatisfaction (S = 0.61), and fear (S = 0.55). The symptoms with the greatest bridge expected influence were anxiety (BEI = 0.92), depression (BEI = 0.80), fear (BEI = 0.52), satiety (BEI = 0.39), appetite (BEI = 0.34), avoided eating out (BEI = 0.24), and limited variety of food (BEI = 0.24).

Discussion: We examined the structure of symptom interrelationships between GI symptoms, ED symptoms, and the psychosocial impact of maintaining a gluten free diet. The results show that pain is the most central symptom in the network, suggesting that pain is a major driver of psychopathology. Overvaluation of weight and shape, a key cognitive symptom of EDs, also has a strong influence on the network and emphasizes the connection between disordered eating and overall symptom severity. Less freedom as a result of a gluten free diet was also highly central, suggesting that quality of life concerns are integral to this symptom network. Anxiety and depression are the strongest bridge symptoms, connecting eating disorder symptoms and GI symptoms. Overall, these findings emphasize that GI clinicians should screen for ARFID and body image concerns in patients with CD.