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

BACKGROUND

Eating disorders represent a complex category of conditions, characterized by dysfunctional thoughts and behaviours about food and body weight or shape, heterogeneous clinical presentations, diagnostic migration and comorbidity. Such complexity highlights the limitations of conventional classification systems, which overlook clinically relevant eating-disorder characteristics such as illness progression. The clinimetric staging model presents an alternative conceptualization that may help differentiate stages of eating-disorder illness by focusing on clinically-relevant but often neglected illness characteristics and comorbidities. A novel statistical approach, namely network analysis, may facilitate the identification of specific eating-disorder stages. Network analysis has been widely applied to model psychiatric conditions as networks of interacting symptoms; this approach is particularly useful for examining the complexity of eating disorders.

METHODS

This study applied network analysis to explore the characteristics of eating disorders across healthy, at-risk, acute, and clinical populations by modelling symptom networks and comparing their structures.

RESULTS

The findings reveal that clinically-significant eating disorder manifestations extend beyond eating-related symptoms and body weight or shape cognitions. In particular, the results underscore the importance of well-being and transdiagnostic dimensions in various populations, excluding clinical groups where symptoms of eating disorders predominate.

CONCLUSIONS

Despite the intricacies of establishing a full staging model for eating disorders, the results obtained provide relevant information regarding clinical manifestations of eating disorders at different levels symptom intensity. This supports not only the development of a more refined staging model for these conditions, but also provides potential prevention and clinical targets for stage-specific interventions.