An approach to sub-grouping the eating disorder population: adding attachment and coping style.

Objective: To investigate whether clinically meaningful sub-groups of patients can be identified by clustering eating disorder features, attachment and coping styles.

Method: 165 patients completed the Eating Disorder Examination (EDE), Attachment Style Questionnaire (ASQ) and Utrecht Coping List (UCL). Cluster analysis was used to identify sub-groups across the sample.

Results: Four clusters were identified. Cluster one had low levels of eating disorder behaviours and the most severe attachment and coping difficulties. Cluster two had high levels of dietary restriction and exercise, and a fearful/avoidant attachment style. Cluster three had high levels of binge eating and vomiting, and few attachment and coping difficulties. Cluster four had low levels of eating disorder features and positive attachment and coping styles.

Conclusions: Clustering participants on the basis of eating disorder features, attachment and coping yields four sub-groups appearing to have clinical face validity.