Narcissistic personality disorder is slated for removal from the next edition of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, to be published in 2013. So notes Charles Zanor in yesterday’s New York Times.

But for some reason, Zanor glossed over the loss of four other personality disorders in the shakeup too — Paranoid, Schizoid, Histrionic and Dependent Personality Disorders. (Schizotypal, Antisocial, Borderline, Avoidant and Obsessive-Compulsive Personality Disorders will remain in the new revision.)

Their intended replacements?

“The Work Group recommends that [these disorders] be represented and diagnosed by a combination of core impairment in personality functioning and specific pathological personality traits, rather than as a specific type.”

Is this a good idea?

The DSM-5 Personality Disorders Workgroup lays out its rationale for the change, with an emphasis on the research showing that there is a significant co-occurrence of personality disorders — that is, people can often meet the criteria for and therefore be diagnosed with more than just one.

The Workgroup also complains that the existing personality disorder categories have arbitrary diagnostic thresholds — but this is an argument that could be made across virtually all of the DSM’s existing diagnostic categories.

The intended hybrid replacement model has not been extensively tested in clinical practice or in practical research. A handful of studies are used to suggest this model is ready for primetime, yet it appears that the Workgroup used a mish-mash of various theories to justify the change.

For instance, they rely on the five-factor model of personality to justify moving to traits. But then discount one of the five factors (openness) as having no significant relationship to personality. Then, like amateur chefs in their first kitchen of personality creation, they add a dash of two more factors not in the five-factor model — compulsivity and schizotypy (a word I’ve never even come across until today!).

I’m sure you can cook something interesting by taking a part of one recipe and throwing in aspects of two other recipes to come up with your own unique dish. And this may be a good model for creativity in a chef.

But in the world of personality and psychological theory, this seems like a very odd and haphazard way to reorganize the diagnostic system for personality disorders that has been the same for nearly three decades.

I’m not alone in thinking this may not be the best idea the DSM-5 folks have ever had:

“They have little appreciation for the damage they could be doing,” [Dr. John Gunderson told the New York Times. …]

“It’s draconian,” he said of the decision, “and the first of its kind, I think, that half of a group of disorders are eliminated by committee.”

He also blamed a so-called dimensional approach, which is a method of diagnosing personality disorders that is new to the DSM. It consists of making an overall, general diagnosis of personality disorder for a given patient, and then selecting particular traits from a long list in order to best describe that specific patient. […]

The dimensional approach has the appeal of ordering à la carte — you get what you want, no more and no less. But it is precisely because of this narrow focus that it has never gained much traction with clinicians.

Indeed, there are some concerns with breaking disordered personality into what seem like fairly arbitrary dimensions — and more of them — complicating the already complex multiaxial system the DSM already uses for diagnosis.

I think Jonathan Shedler, a psychologist at the University of Colorado’s Medical School, hit the nail on the hear with this quote:

“Clinicians are accustomed to thinking in terms of syndromes, not deconstructed trait ratings. Researchers think in terms of variables, and there’s just a huge schism.”

He said the committee was stacked “with a lot of academic researchers who really don’t do a lot of clinical work. We’re seeing yet another manifestation of what’s called in psychology the science-practice schism.”

There is an ongoing disconnect between researchers — who rarely engage in clinical practice — and clinicians — who actually have to use the researchers’ categories and paradigms in daily practice.

Of course, the DSM-5 folks suggest their workgroups have equal and adequate representation of all parties on them. Yet this is a stinging example of where it seems that the clinician’s point of view is simply not being heard.

While practice should not rule out good science, good science should also take into account good practice and what’s done in the real world. Foisting a new trait-based system onto clinicians while removed half of the existing personality disorders from the new edition is likely to cause more problems than it solves.

Read the full article: Narcissistic Disorder to Be Eliminated in Diagnostic Manual