Three-year symptom trajectories and prognostic factors in eating disorders: the prospective EDILS cohort study

Background: Eating disorders (EDs) are severe psychiatric conditions characterized by a chronic and often fluctuating course. Identifying factors influencing symptom trajectories is essential to improve patient management and long-term outcomes. The objectives are to examine symptom evolution and identify prognostic factors across the three main ED subtypes, restrictive, bulimic, and compulsive disorders, in patients initiating care in a hospital nutrition department. Methods: Patients newly referred to a university hospital nutrition department were prospectively followed for three years within the EDILS cohort. ED symptom severity was assessed using validated self-report instruments, including the Eating Disorder Inventory-2 (EDI-2), alongside measures of anxiety, depression, body dissatisfaction, and sociodemographic and behavioral variables such as smoking. Longitudinal analyses were conducted to evaluate symptom trajectories and associated predictors over time. Results: The cohort included 379 patients (87.6% women; mean age 36.0 ± 13.5 years) with binge-eating (n=222), restrictive (n=108), and bulimic (n=49) disorders. Baseline symptom severity was highest in bulimic patients, who also showed greater symptom improvement over three years than restrictive and binge-eating profiles (mean EDI-2 change −23.5 vs −12.1 and −15.2; p=0.04). Older age and smoking at baseline were associated with symptom worsening at 36 months. Longitudinal analyses further indicated that persistent smoking, anxiety, depression, body dissatisfaction, and lower family life satisfaction were independently associated with less favorable symptom trajectories Conclusion: ED symptoms improved overall over three years with symptom trajectories were mainly influenced by psychiatric comorbidities, body dissatisfaction, smoking, and relational factors, while older age and smoking predicted symptom worsening. These findings highlight the importance of targeting modifiable psychological, behavioral, and social factors to support personalized, integrated care in eating disorders.