Exploring neuropsychological and socio-emotional performance in anorexia nervosa: A cluster analytic approach

Abstract

Objective: The aim of this study was to characterise heterogeneity in neuropsychological and socio-emotional test performance in young women with anorexia nervosa (AN) using hierarchical cluster analysis. Further, we aimed to test whether cognitive profiles were associated with differences in clinical variables (BMI, illness duration, and age at diagnosis), psychopathology (eating disorder, autistic symptoms, anxiety, and depression), and functional impairment. Method: set-shifting, central coherence, and theory of mind abilities were measured in 118 participants with acute or remitted AN. A hierarchical cluster analysis using Ward’s method with a Euclidean distance measure was performed with the neuropsychological and socio-emotional variables. Differences between clusters were assessed using ANOVAs. Results: four clusters emerged, with significant differences in neuropsychological and socio-emotional functioning. There were no significant differences between clusters in clinical variables, psychopathology, or functional impairment, however these analyses lacked power due to small cluster sizes. Conclusions: our results demonstrate significant heterogeneity in cognitive profiles in AN, supporting a more personalised approach to treatment. Studies in larger samples are required to establish whether these variables map onto clinically significant differences in aetiology, clinical presentation, comorbidity patterns, and/or treatment responses.