The aim of the present study was to analyse outcome, time to remission, and predictors of time to remission in a cohort of Danish eating disorder patients. Seventy-eight patients (35 anorexic, 30 bulimic and 13 unspecified eating disorder patients) were interviewed 2 years after initial assessment. Method of assessment was Eating Disorder Examination (EDE), Longitudinal Interval Follow-up Evaluation of Eating Disorders (LIFE-EAT-II), Eating Disorder Inventory (EDI), Symptom Check List (SCL-90R), Present State Examination (PSE) and the Structured Clinical Interview for DSM-III-R Axis-II (SCID-II). Method of analysis was Kaplan-Meier estimate of survival, Log Rank test and Cox regression analysis. In total 48.7% reached remission with mean time to remission at 27 months. A trend difference between the diagnostic groups when measuring time to remission was found, i.e. patients with unspecified eating disorders remitted faster than bulimic (BN) patients who in turn remitted faster than anorexic (AN) patients. Body mass index (BMI) at baseline was the best predictor of time to remission for the total sample. Predictors differed when looking at diagnostic groups separately. Final outcome was comparable with earlier studies while relapse frequency was low. Patients with AN remitted faster than found in earlier survival analysis studies, while the remission rate for BN patients was comparable with earlier studies. Despite the prognostic value of BMI for the total sample, predictor analysis implied more disorder diversity than homogeneity.