Network analysis models of eating disorders (NAMED): an exploratory study of symptoms configuration at different stages of illness

Eating disorders (ED) represent a complex category of psychiatric conditions, characterized by dysfunctional thoughts and behaviours about food and excessive preoccupation with body weight or shape, heterogeneous presentations, diagnostic migration and comorbidity. Such complexity underlined the limits on conventional diagnostic classification systems that do not include trajectories of symptoms and possible comorbidities. An alternative conceptualization of disease is the clinimetric staging model, which aims to determine the current stage of a psychiatric disorder, ranging from prodromal to chronic, by focusing on the characteristics of a condition at a particular point in time. The staging model helps differentiate between the progression of clinical manifestations. Recently network analysis (NA) has been widely applied to explore ED at the symptom level, in an attempt to tackle its intricacies. NA allows to model psychiatric conditions as a constellation of interacting symptoms and appears particularly relevant to explore the complexity of ED. The following explored the characteristics of ED at different stages, modelling symptom networks and comparing their structures. The results highlight the central role of positive functioning and transdiagnostic dimensions in all but the clinical population, where ED symptoms were more prominent. These results support the idea that ED is a complex category of conditions in which clinically meaningful manifestations reach beyond the symptomatology strictly related to eating behaviours and cognitions. While it was past the scope of this work to establish a full staging model for ED, the results obtained provide relevant information regarding clinical manifestations of ED at different levels of ED symptom intensity, thus supporting not only the development of a more refined staging model for these conditions but also providing potential clinical targets for stage-specific interventions.