News of the WeekPsychiatry

APA Seeks to Overhaul Personality Disorder Diagnoses

See allHide authors and affiliations

Science  12 Mar 2010:
Vol. 327, Issue 5971, pp. 1314
DOI: 10.1126/science.327.5971.1314

This article has a correction. Please see:

Personality disorders are hard to pin down. They don't have a common defining mood or behavior, people don't get hospitalized for having one, and a drug won't cure one. But they can cause all kinds of havoc. People with antisocial personalities, regularly encountered in crime news, are well-known. But other types of these disorders feed into high rates of alcoholism, drug addiction, suicide, and the dysfunction found in many offices, families, or soap operas.

A personality disorder “at the core” involves “failure to develop healthy functioning in self and interpersonal domains,” according to psychologist Lee Anna Clark of the University of Iowa in Iowa City, a member of the American Psychiatric Association's (APA's) work group that is proposing revisions for personality disorders in the forthcoming fifth version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V), often referred to as psychiatry's bible (Science, 12 February, p. 770). These disorders are in for a “major overhaul,” says APA psychiatrist Darrell Regier, co-chair of the broader effort. The goal is to create a more flexible and precise system, but some critics say the changes are too complicated for the busy clinician.

The current edition, DSM-IV, identifies 10 personality disorders, with a list of criteria for each. But these have been criticized for having arbitrary diagnostic thresholds and for clumping together people with quite different symptoms. Work group chair Andrew Skodol, a psychologist at the University of New Mexico, Albuquerque, says, for example, that with nine criteria (five required for a diagnosis), there are 256 ways that a person can be diagnosed with “borderline” personality disorder. Furthermore, “more than half of people diagnosed with one personality disorder are diagnosed with a second or third or even fourth,” says Clark. And despite all the choices, many end up in the “not otherwise specified” (NOS) category. The work group is proposing a system members feel will better capture the range of pathology.

Take borderline personality disorder. “Borderline” used to be a way of referring to a state somewhere between neurotic and psychotic. But since the 1970s, it has become one of the more extensively researched personality types—often found in young women who tend to be self-centered, hysterical, emotionally dependent, and often suicidal. Marilyn Monroe is often cited as an example; another is Glenn Close's character in Fatal Attraction.

Slimmed down.

Proposed changes would reduce the number of personality “types.”

Under DSM-IV, a clinician would make a diagnosis of borderline personality disorder if a patient displayed at least five from a list of nine symptoms such as impulsiveness, self-mutilating behavior, and “chronic feelings of emptiness.” Under the new regime, the clinician would first determine whether the person's functioning is so impaired that she qualifies for a general diagnosis of personality disorder. The heart of the assessment would be a “trait profile” compiled from a list of six major domains of personality covering areas such as “introversion,” “antagonism,” and “schizotypy” (odd or eccentric behavior).

The clinician could also refer to five suggested “types” of personality disorders—including “borderline type”—pared down from the current 10 (see box). The four others would be antisocial/psychopathic, avoidant (anxious and fearful of rejection), obsessive-compulsive, and schizotypal. The latter two contain elements of more serious diagnoses in the categories of anxiety and psychosis but do not meet all the criteria.

One venerable type did not make the cut: narcissistic personality disorder. That's controversial, says psychologist Drew Westen of Emory University in Atlanta. The diagnosis is “widely used—and nobody has suggested it doesn't exist.” Skodol says narcissism is a trait seen in more than one type of disorder and is covered under the domain of “antagonism.”

On the edge.

Alex Forrest, played by Glenn Close in Fatal Attraction, fits the diagnosis of a borderline personality.


Clark and Skodol say that in practice, many people with personality disorders will fall outside those five types. But they'll get better diagnoses because, instead of being designated “NOS,” they will be profiled based on the six traits.

Some experts question how useful the new system will be. Westen lauds the group for trying to make diagnosis “more sophisticated clinically.” But, he says, by trying to combine the list of five personality types with six crosscutting traits, “what they've come up with is a camel—a horse made by committee.”

Psychologist Thomas Widiger of the University of Kentucky in Lexington thinks the proposed overhaul is “terrible,” leaves too much to the clinician's subjective judgment, and will create new headaches for insurance companies. It is “so complicated, it will not be used,” he says. “Clinicians will match to one of those five [types]. Then they'll be done.” Skodol is more optimistic, saying that clinicians can choose how deeply detailed their diagnosis is based on “available time, information, and expertise.”

Skodol's group also wants to put personality disorders on the same footing as other mental disorders. DSM lists disorders in two different categories, or “axes.” Most are on Axis I; personality disorders are on Axis II (along with intellectual disabilities) because they are seen as stable, chronic conditions rather than episodic illnesses. Moving them to Axis I, says Clark, “makes the statement that personality disorder is just as critical to assess as any other diagnosis.”

View Abstract

Stay Connected to Science

Navigate This Article