King's College London

Research portal

Neuro- and social-cognitive clustering highlights distinct profiles in adults with anorexia nervosa

Research output: Contribution to journalArticle

Bethany Renwick, Peter Musiat, Anna Lose, Hannah Dejong, Hannah Broadbent, Martha Kenyon, Rachel Loomes, Charlotte Watson, Shreena Ghelani, Lucy Serpell, Lorna Richards, Eric Johnson-Sabine, Nicky Boughton, Janet Treasure, Ulrike Schmidt

Original languageEnglish
Pages (from-to)26-34
Number of pages9
JournalThe International journal of eating disorders
Issue number1
Early online date3 Nov 2014
E-pub ahead of print3 Nov 2014
Published1 Jan 2015

Bibliographical note

© 2014 Wiley Periodicals, Inc. (Int J Eat Disord 2015; 48:26–34)


King's Authors


OBJECTIVE: This study aimed to explore the neuro- and social-cognitive profile of a consecutive series of adult outpatients with anorexia nervosa (AN) when compared with widely available age and gender matched historical control data. The relationship between performance profiles, clinical characteristics, service utilization, and treatment adherence was also investigated.

METHOD: Consecutively recruited outpatients with a broad diagnosis of AN (restricting subtype AN-R: n = 44, binge-purge subtype AN-BP: n = 33 or Eating Disorder Not Otherwise Specified-AN subtype EDNOS-AN: n = 23) completed a comprehensive set of neurocognitive (set-shifting, central coherence) and social-cognitive measures (Emotional Theory of Mind). Data were subjected to hierarchical cluster analysis and a discriminant function analysis.

RESULTS: Three separate, meaningful clusters emerged. Cluster 1 (n = 45) showed overall average to high average neuro- and social- cognitive performance, Cluster 2 (n = 38) showed mixed performance characterized by distinct strengths and weaknesses, and Cluster 3 (n = 17) showed poor overall performance (Autism Spectrum disorder (ASD) like cluster). The three clusters did not differ in terms of eating disorder symptoms, comorbid features or service utilization and treatment adherence. A discriminant function analysis confirmed that the clusters were best characterized by performance in perseveration and set-shifting measures.

DISCUSSION: The findings suggest that considerable neuro- and social-cognitive heterogeneity exists in patients with AN, with a subset showing ASD-like features. The value of this method of profiling in predicting longer term patient outcomes and in guiding development of etiologically targeted treatments remains to be seen.

Download statistics

No data available

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454