Eating disorder symptom non-endorsers in hospitalised patients with anorexia nervosa: Who are they?

ObjectiveImpaired insight and illness denial are common in anorexia nervosa (AN). Missing an AN diagnosis may delay treatment and negatively impact outcomes.MethodThe current retrospective study examined the prevalence and characteristics of AN symptom non-endorsement (i.e., scoring within the normal range on the Eating Disorder Examination Questionnaire [EDE-Q] or the Eating Disorder Examination [EDE] interview) in three independent samples of hospitalised patients with AN (N1 = 154; N2 = 300; N3 = 194). A qualitative chart review of a subsample of non-endorsers (N4 = 32) extracted reports of disordered eating behaviours observed by the treatment team.ResultsThe prevalence of non-endorsement ranged from 11% to 34% across sites. Non-endorsers were more likely to be diagnosed with AN restricting type (AN-R) and reported fewer symptoms of co-occurring psychopathology than endorsers. Groups benefitted equally from treatment. The qualitative chart review indicated that objective symptoms of AN were recorded by staff in over 90% of non-endorsers.ConclusionsEating disorder symptom assessments using the EDE-Q or EDE may miss symptomatology in up to a third of individuals hospitalised with AN. This study highlights the potential utility of multi-modal assessment including patient interviews, collateral informants, and behavioural observation to circumvent non-endorsement. Up to one-third of patients with AN ill enough to be hospitalised scored within the normal range on eating disorder symptom questionnaire and interview. Symptom non-endorsement was most common in those with restricting type of AN Multi-modal assessment including patient interviews, collateral informants and behavioural observation may be needed to properly assess AN symptoms.