Dimensional versus categorical classification of prototypic and nonprototypic cases of personality disorder

This study examined interrater reliability and ratings of confidence and clinical utility (professional communication, case conceptualization, treatment planning) of categorical and dimensional approaches to diagnosing prototypic and nonprototypic personality disorder cases. Two national samples of psychologists (n = 93, n = 92) participated. Interrater reliability was higher for prototypic cases than nonprototypic cases for the categorical system, but similar for prototypic and nonprototypic cases using dimensional ratings. Across cases, interrater reliability and confidence were highest for the categorical model, hybrid models, and the five-factor model. However, ratings of clinical utility were highest for the categorical and the hybrid models, even when interrater reliability was inadequate, suggesting clinician preference for a classification based on the existing categories. Mean ratings for the prototypic cases supported the theorized relationships between the dimensional models and the personality disorders. Reasons for these findings and implications for moving towards a dimensional model of personality disorder are discussed.