Personality subtypes in adolescents with eating disorders: validation of a classification approach


  • Conflict of interest statement: No conflicts declared.

  • 1

    The SWAP 200 Q-sort includes 200 comprehensive items reflective of all aspects of personality functioning and pathology, and directs the assessor to sort items into a fixed distribution that facilitates statistical aggregation. The Q-factor method takes advantage of the Q-sort methodology to aggregate subjects rather than items into groups.

Heather Thompson-Brenner, Center for Anxiety and Related Disorders, Department of Psychology, Boston University, 648 Beacon St., Boston MA 02215, USA; Fax: 617 353 9609; Email:


Background:  Research has identified three personality subtypes in adults with eating disorders (EDs): a high-functioning, an undercontrolled, and an overcontrolled group. The current study investigated whether similar personality prototypes exist in adolescents with EDs, and whether these personality prototypes show relationships to external correlates indicative of diagnostic validity.

Methods:  Experienced clinicians from an adolescent practice-research network provided data on ED symptoms, DSM-IV comorbidity, personality pathology, and family and developmental history for 120 adolescent patients with EDs.

Results:  Consistent with the findings from the adult literature, three types of personality pathology emerged in adolescents: High-functioning/Perfectionistic, Emotionally Dysregulated, and Avoidant/Depressed. The High-functioning prototype showed negative associations with comorbidity and positive associations with treatment response. The Emotionally Dysregulated prototype was specifically associated with externalizing Axis I and Cluster B Axis II disorders, poor school functioning, and adverse events in childhood. The Avoidant/Depressed prototype showed specific associations with internalizing Axis I and Clusters A Axis II disorders, poor peer relationships, poor maternal relationships, and internalizing disorders in first-degree relatives.

Conclusions:  These data support the presence of at least three diagnostically meaningful personality prototypes in adolescents with EDs, similar to those found previously in adults. Diagnosis of adolescents with EDs may be usefully supplemented by the assessment of personality style.