This Month’s Expert: Diagnosing Personality Disorders Michael B. First, M.D.

This Month’s Expert: Diagnosing Personality Disorders Michael B. First, M.D.TCPR: Dr. First, you edited DSM-IV and were the lead author of SCID, the Structured Clinical Interview for DSM-IV disorders. You literally wrote the book on psychiatric diagnosis. I have a lot of questions about personality disorders. To start, why were personality disorders separated from other disorders and put into Axis-II?

Dr. First: This split between Axis I and II came about in 1980 with the publication of DSM-III. It came out of a concern that disorders like personality disorders, learning disorders, mental retardation and autism could be overlooked because other, more florid disorders would take everyone’s attention. So this was strictly a practical mechanism, to keep these disorders from being ignored, and there was no conceptual basis for this separation at all.

TCPR: How has the Axis I vs. Axis II system worked out over the years?

Dr. First: On the plus side, it has given these disorders more prominence, both among clinicians and researchers. Having a separate Axis for personality disorders forces anyone doing a multiaxial evaluation to at least consider whether or not there is a personality disorder present for each patient. But the down side is that it created this illusion that somehow things on Axis II are different animals from things on Axis I. Various myths have developed around Axis II conditions: that Axis II disorders are not treatable, or that Axis I disorders are treatable with medications but Axis II are treatable with therapy only.

TCPR: But aren’t personality disorders enduring, long-lasting traits, as opposed to Axis I disorders, which are supposed to be more discrete and episodic?

Dr. First: It’s true that people have tried to come up with some conceptual way of separating the two Axes, saying that personality disorders are more stable and more trait-like, but in fact many Axis I disorders are chronic, like schizophrenia, or paraphilia, or even chronic depression. Furthermore, Axis II disorders are much less stable than we had originally thought. There really are no clear conceptual distinctions. And, there is a dark side of this system, which is that it may help insurance companies discriminate against the treatment of Axis II; they can justify this by pointing to the DSM, saying that since the APA considers the disorders different, they don’t have to cover them.

TCPR: So what do you think is likely to happen in DSM-V?

This article originally appeared in The Carlat Psychiatry Report -- an unbiased monthly covering all things psychiatry.
Want more, plus easy CME credit?
Subscribe today!

Dr. First: It’s likely that personality disorders will be on Axis I. Moving disorders form Axis II to Axis I has happened before. When we went from DSM-III to III-R, the autism community wanted autism to be on Axis I, and so it was moved. Then, when we went from DSM III-R to DSM-IV, the learning disorders community wanted learning disorders to go to Axis I, and we made that change. But the mental retardation community still liked the idea that MR was special, and they wanted it to stay on Axis II. Now, most clinicians and researchers in personality disorders are saying it is time to move PDs to Axis I.

TCPR: So what would remain on Axis II in this scenario?

Dr. First: MR might remain, which makes sense, because having MR on Axis II is basically a way of saying that when evaluating children, it is very important to know their level of intellectual functioning regardless of their Axis I disorder. In addition, personality traits, as opposed to disorders, could remain on II. After all, it is useful to know a patient’s personality traits regardless of their Axis I diagnosis, because it can help you choose the best way to approach treating the Axis I disorder with that particular patient. Even if someone doesn’t have any mental disorder, being aware of their personality traits is useful. For example, if you are on a CL service and you are consulting on a medical patient, you might find that they have no DSM disorder but that they have obsessive compulsive traits, and that might be useful information for the medical team in order to more effectively communicate with that patient about various treatment options.

TCPR: It seems that psychiatrists are diagnosing personality disorders less these days.

Dr. First: I think that’s true, and that partly reflects psychiatry’s increasing emphasis on psychopharmacology. From a purely pragmatic perspective, one reason for writing down a diagnosis is to justify the medication you are using. This is certainly the way many psychiatrists use the DSM. But in fact there is a body of literature that shows that when patients have a comorbid personality disorder, it’s more difficult to treat their Axis I conditions. For example, if you have a patient who is depressed and who has narcissistic or borderline PD, that person is likely going to be difficult to treat. They may be “failing” various trials of meds, and knowing about their PD would be very helpful in understanding why. Or, a patient with panic disorder may not be responding to medication because they have a comorbid avoidant personality disorder. This is one reason drug companies exclude such patients from clinical trials, because they want the best response rate they can get.

TCPR: One of the controversies in the field is the distinction between bipolar disorder and borderline personality disorder.

Dr. First: This is a result of the trend to view bipolar disorder as a spectrum. Bipolar disorder is traditionally defined in terms of discrete episodes, but as you stretch that definition out and include mixed episodes, irritability, and brief mood swings, it can look a lot like borderline PD. Certainly, if you look at some of the criteria for borderline PD, they can sound like bipolar disorder. For example, the item referring to excessive and inappropriate anger could also apply to people with bipolar disorder who have irritable manic episodes. The impulsivity item resembles some typical reckless manic behaviors. And the item “chronic feelings of emptiness” resembles the chronic anhedonia seen in depression, which can be part of the bipolar picture. But then there are also some traits that are clearly more characteristic of the borderline PD picture, such as fear of abandonment or idealization/devaluation.

TCPR: Even idealization/devaluation can seem to occur in bipolar disorder, in the sense that when you are depressed everything and everybody looks terrible and when you are manic, everything and everybody looks great.

Dr. First: That’s true. It depends on how you look at the items. But generally, you can distinguish bipolar disorder because of its more episodic nature and the fact that in borderline personality disorder, idealization/devaluation only occurs with individuals who are emotionally very important to the patient. It should be noted that there are some experts in the field, such as Hagop Akiskal, who argue that borderline PD doesn’t really exist and is just part of the bipolar spectrum.

TCPR: You spend much of your time teaching interviewing skills. What are some tips for how to assess personality traits?

Dr. First: Although I’m one of the authors of the SCID-II, which is used by researchers to diagnose PDs, I know that it is not very practical to use in standard clinical practices. It requires a large amount of insight on the part of patients, because it goes through the criteria, one by one, asking patients to judge how they have behaved over the years. In practice, the best way to diagnose personality disorders is with an open-ended interview in which you look for clues regarding their personality based on the answers to questions about their relationship history and their work history, and closely observe how the patient is behaving during the interview.

TCPR: That’s a lot of information to get in one session.

Dr. First: You may not be able to come to a conclusion in a single session. And you may very well need to talk to other informants, because patients often have a hard time seeing their own traits accurately. I’ll ask things of the informant like, “How do you see your husband? What kinds of things does he do that are annoying?” A patient might come in and say that he is well-respected at work, but when you talk to the wife about his work relationships, she might respond instead that “all of his colleagues complain that he is very hard to work with and is arrogant.” This might suggest the presence of narcissistic traits of which the patient may be totally unaware.

TCPR: What kinds of behaviors during the interview are useful to notice?

Dr. First: Someone who appears shy and meek raises a red flag for either dependent or avoidant PD, someone who is cagey and suspicious of everything you say might suggest paranoid or schizotypal PD, a rigid personality implies obsessive compulsive PD, arrogance suggests narcissistic PD, and someone who is overly familiar suggests histrionic PD. I should also mention that asking about childhood abuse is important. While not everyone who has been abused has a PD, there is a significant correlation.

This Month’s Expert: Diagnosing Personality Disorders Michael B. First, M.D.

This article originally appeared in:

The Carlat Psychiatry Report
Click on the image to learn more or subscribe today!

This article was published in print 9/2008 in Volume:Issue 6:9.


APA Reference
First,, M. (2013). This Month’s Expert: Diagnosing Personality Disorders Michael B. First, M.D.. Psych Central. Retrieved on September 18, 2020, from


Scientifically Reviewed
Last updated: 7 Sep 2013
Last reviewed: By John M. Grohol, Psy.D. on 7 Sep 2013
Published on All rights reserved.