An empirically derived classification of adolescent personality disorders.

ABSTRACT

OBJECTIVE: This study describes an empirically derived approach to diagnosing adolescent personality pathology that is clinically relevant and empirically grounded. METHOD: A random national sample of psychiatrists and clinical psychologists (N = 950) described a randomly selected adolescent patient (aged 13-18 years, stratified by age and gender) in their care using the Shedler-Westen Assessment Procedure-II-A for Adolescents (SWAP-II-A) and several additional questionnaires. RESULTS: We applied a form of factor analysis to identify naturally occurring personality groupings within the patient sample. The analysis yielded 10 clinically coherent adolescent personality descriptions organized into 3 higher-order clusters (internalizing, externalizing, and borderline-dysregulated). We also obtained a higher-order personality strengths factor. These factors and clusters strongly resembled but were not identical to factors similarly identified in adult patients. In a second, independent sample from an intensive day treatment facility, 2 clinicians (the patients' treating clinician and the medical director) independently completed the SWAP-II-A, the Child Behavior Checklist (CBCL), and a measure of adaptive functioning. Two additional clinicians, blinded to the data from the first 2 clinicians, independently rated patients' ward behavior using a validated measure of interpersonal behavior. Clinicians diagnosed the personality syndromes with high agreement and minimal comorbidity among diagnoses, and SWAP-II-A descriptions strongly correlated in expected ways with the CBCL, adaptive functioning, and ward ratings. CONCLUSION: The results support the importance of personality diagnosis in adolescents and provide an approach to diagnosing adolescent personality that is empirically based and clinically useful.