New way to classify personality disorders proposed

The study by Christopher Hopwood and colleagues calls for a more scientific and practical method of categorizing personality disorders -- a proposal that ultimately could improve treatment, Hopwood said.

"We're proposing a different way of thinking about personality and personality disorders," said Hopwood, MSU assistant professor of psychology and an experienced clinician. "There's widespread agreement among personality disorder researchers that the current way to conceptualize personality disorders is not working."

The study is being cited by the team of experts that currently is developing criteria for the manual used to diagnose personality disorders -- the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders, or DSM-5, slated to come out in 2013.

The study is being considered for inclusion in the DSM-5. The DSM, published by the American Psychiatric Association, is considered the bible of the U.S. mental health industry and is used by insurance companies as the basis for treatment approval and payment. The study also will appear in an upcoming issue of the Journal of Personality Disorders.

The current method of classifying personality disorders, as spelled out in the fourth edition of the DSM, or DSM-IV, breaks personality disorders into 10 categories, Hopwood said. That system is flawed, he said, because it does not take into account severity of personality disorders in an efficient manner and often leads to overlapping diagnoses.

"It's just not true that there are 10 types of personalities disorders, and that they're all categorical -- that you either have this personality disorder or you don't," Hopwood said. "Scientifically, it's just not true."

Hopwood and colleagues propose a new three-stage strategy for diagnosing personality disorders:

Stage One: Consider a patient's normal personality traits, such as introversion/extroversion. "If a person is depressed and I'm a clinician, it might make a difference if I think they're extroverted depressive rather than introverted depressive," Hopwood said. "It may dictate the type of recommendations I make for them." These normal personality traits also may indicate patient strengths that could help in overcoming psychiatric difficulties; such strengths are not assessed in the current DSM.

Stage Two: Create a numerical score to represent severity of the disorder. "We're arguing that one single score can represent that severity, so clinicians can easily communicate with one another about how severe a patient is," Hopwood said. "That may indicate decisions such as whether this person should be hospitalized or treated with outpatient care."

Stage Three: Condense the list of 10 personality disorder categories to five dimensional ratings. Under this proposal, clinicians would diagnose how many symptoms of each disorder a patient has, rather than whether they have one or more of 10 disorders as in the current system. Hopwood said this is more reliable, valid and specific than the current system. He added that research has not sufficiently supported the validity of several current personality disorders. The proposed dimensional ratings are:

• Peculiarity. The defining characteristic here is oddness in thought or behavior. This dimension includes the diagnoses of paranoid, schizotypal and schizoid.
• Withdrawal. This includes avoidant personalities. "This may have to do with not wanting to leave the house," Hopwood said.
• Fearfulness. This combines disorders with opposite extremes of harm avoidance, such as antisocial (which involves fearlessness) and dependant or avoidant (which involves fearfulness).
• Unstable. This is similar to the diagnosis of borderline in DSM-IV. The defining characteristic is instability, such as with relationships, identity or emotional experience.
• Deliberate. This includes obsessive-compulsive disorder and other disorders defined by overly methodical behavior. "It's having a rigid sense of how life should happen -- how I should behave and how other people should behave," Hopwood said.

Ultimately, Hopwood said, the proposal could improve both the system for diagnosing personality disorders as well as the outcome. "Presumably, if this leads to better clinical efficiency it could lead to better clinical care, and that's in everybody's interest," he said.

Co-authors of the proposal include Andrew Skodol of the Sunbelt Collaborative, New York State Psychiatric Institute and Columbia Medical School, and Leslie Morey of Texas A&M University.