Bret S. Stetka, MD; Christoph U. Correll, MD


May 21, 2013

In This Article

Personality Disorders

The Change

In developing diagnostic criteria for personality disorders, the DSM-5 Work Group initially proposed a somewhat dramatic new approach: Maintain 6 personality disorder diagnoses from the prior 10 in DSM-IV, and move from a categorical to a trait-based, dimensional classification system. Per the categorical system, a patient either has a diagnosis or not, whereas a dimensional system better captures the nuances of human personality by measuring a variety of traits on a continuum. The proposal was ultimately voted down; however, the alternative hybrid dimensional-categorical model is included in a separate chapter in Section 3 of DSM-5 to stimulate further research on this modified classification system.

Of note, despite the required enduring and impairing nature of personality disorder symptoms and traits, in the field trials, only borderline personality disorder had good interrater reliability. In contrast, obsessive-compulsive personality disorder and antisocial personality disorder were in the questionable reliability range, and too few patients with other personality disorders were included to test their reliability. Although all original 10 personality disorders from DSM-IV were finally retained, DSM-5 has moved from the multiaxial to a monoaxial system that removes the arbitrary boundaries between personality disorders and other mental disorders.

The Implications

The integration of personality disorders with other psychiatric diagnoses that were previously separated and classified on a different axis returns to a more unified and dimensional view of personality, character, temperament, and mental illness. On the other hand, retention of personality disorders with questionable or untested reliability can lead to the perpetuation of stigmatizing diagnoses that may lack validity and for which treatment approaches are also often unclear. Moreover, too few patients in the field trials were available to test 7 of the 10 personality disorders, suggesting that future field trials should urgently include community mental health centers and clinicians working in nonacademic private settings, where these patients are more likely to be found and treated -- often with psychotherapy.

However, the fact that borderline personality disorder had such good interrater reliability, whereas the other personality disorders did not, may support previously endorsed views that it could belong in the bipolar disorder spectrum rather than being classified as a personality disorder. Future research will hopefully help clarify this further.