The Corpus Callosum

The American Psychiatric Association is considering whether to href="">reboot
their diagnosis machine.  In 1952, the Diagnostic and Statistical
Manual (DSM) was published.  In 1980, the third edition was
published.  The third edition was important, because for the first
time, it required the use of specific criteria for establishing a
diagnosis.  (See href="">this
New Yorker article for a description of the history of DSM and the
development of the descriptive approach.)

There have been many criticisms of this approach.  Google
criticisms of DSM if you are curious.  Most of the criticisms are
not very interesting, because they tend to be be based upon the truism
that all models are false.  The DSM is a model, and it is,
accordingly, false.  True, but uninteresting.  The real
question: is it useful?  To some extent, that depends upon how it
is used.  If it is used incorrectly, problems can arise. 
Some arguments put forth as criticisms of the DSM actually are
criticisms of how the DSM is used, not the document itself.  Many
of these arguments have merit, but that is a different subject. I won’t
get into that today.

The changes being proposed are not intended to make the model more
true.  Rather, the changes are hoped to make it more useful. 

To clarify, a lot of changes are being proposed, but this post will
contain a discussion of only one of the proposed changes.  That
is, it has been proposed to change the way that personality disorders
are established.  Currently, this is a two-step process.  The
first step is to establish whether the general criteria for a
personality disorder are met. 

diagnostic criteria for a Personality Disorder

A. An enduring pattern of inner experience and behavior that deviates
markedly from the expectations of the individual’s culture. This
pattern is manifested in two (or more) of the following areas:

    (1) cognition (i.e., ways of perceiving and
interpreting self, other people, and events)
    (2) affectivity (i.e., the range, intensity,
lability, and appropriateness of emotional response)
    (3) interpersonal functioning
    (4) impulse control

B. The enduring pattern is inflexible and pervasive across a broad
range of personal and social situations.

C. The enduring pattern leads to clinically significant distress or
impairment in social, occupational, or other important areas of

D. The pattern is stable and of long duration and its onset can be
traced back at least to adolescence or early adulthood.

E. The enduring pattern is not better accounted for as a manifestation
or consequence of another mental disorder.

F. The enduring pattern is not due to the direct physiological effects
of a substance (e.g., a drug of abuse, a medication) or a general
medical condition (e.g., head trauma).

If those criteria are met, then the second step comes into play. 
The second step is a determination of the applicability of one or more
of the criterion sets for a specific personality disorder.  For
example, here are the criteria for Paranoid Personality Disorder:

criteria for 301.0 Paranoid Personality Disorder

A. A pervasive distrust and suspiciousness of others such that their
motives are interpreted as malevolent, beginning by early adulthood and
present in a variety of contexts, as indicated by four (or more) of the

    (1) suspects, without sufficient basis, that others
are exploiting, harming, or deceiving him or her
    (2) is preoccupied with unjustified doubts about the
loyalty or trustworthiness of friends or associates
    (3) is reluctant to confide in others because of
unwarranted fear that the information will be used maliciously against
him or her
    (4) reads hidden demeaning or threatening meanings
into benign remarks or events
    (5) persistently bears grudges, i.e., is unforgiving
of insults, injuries, or slights
    (6) perceives attacks on his or her character or
reputation that are not apparent to others and is quick to react
angrily or to counterattack
    (7) has recurrent suspicions, without justification,
regarding fidelity of spouse or sexual partner

B. Does not occur exclusively during the course of Schizophrenia, a
Mood Disorder With Psychotic Features, or another Psychotic Disorder
and is not due to the direct physiological effects of a general medical

Note: If criteria are met prior to the onset of Schizophrenia, add
“Premorbid,” e.g., “Paranoid Personality Disorder (Premorbid).”

As an aside, I often have seen students or untrained, or improperly
trained, persons jump directly to the specific criterion sets, without
first establishing the applicability of the general criteria. 
That is a mistake.

Back to the matter at hand.  The proposed change would get rid of
the criteria sets, and instead call for the use of a system that rates
persons on a variety of scales.  The same scales would be used for
each disorder.  This change can be conceptualized as a change from
a categorical to a dimensional system.  If
implemented, it would be a fundamental shift in the methodology used to
diagnose personality disorders.

Unfortunately, the article I’m gong to discuss is not open
access.  (I am breaking one of my rules here.)  The abstract
is openly available, as is an editorial about the article.

Clinicians Recognize DSM-IV Personality Disorders From Five-Factor
Model Descriptions of Patient Cases?

Future of Personality Disorders in DSM-V?

Here’s a snippet from the introduction to the article.  This
encapsulates the crux of the issue and the proposed change:

The Diagnostic and Statistical Manual of Mental Disorders
(DSM) is under revision. One proposal for the pending DSM-V is
dimensionalizing personality disorders, and the Five-Factor Model (FFM)
has received the most attention, either as a supplement or replacement
for axis II. Whereas the DSM-IV classifies maladaptive personality with
10 discrete disorders defined by unique criteria, the FFM describes
personality in a continuous manner along 30 traits grouped into five
factors (Figure 1) identified as reflecting the bulk of the variance
among personalities. The FFM is a promising candidate for the DSM-V
because it has been shown to be biologically based, universal,
temporally stable, and can avoid problems with the DSM-IV axis II
categories including high comorbidity and arbitrary diagnostic

So how would this be more useful?

The various criterion sets appear, at first glance, to describe
reasonably discrete conditions.  However, in actual practice,
there is considerable overlap.  The result is that many patients
end up with fuzzy labels.  For example, you may see a diagnosis
listed as “mixed personality disorder with histrionic and borderline
features,” or “Cluster B personality traits” or some such weaselly
construct.  The editorial, linked above, informs us that the most
common personality disorder diagnosis is “personality disorder not
otherwise specified (PDNOS).”

The limitations of DSM categorical conceptualizations of
personality disorders are well known: excessive co-occurrence among
disorders, extreme heterogeneity among patients receiving the same
diagnosis, arbitrary diagnostic thresholds for the boundaries between
pathological and “normal” personality functioning, and inadequate
coverage of personality psychopathology such that the diagnosis of
personality disorder not otherwise specified (PDNOS) is the most common.

For a variety of purposes to which a diagnosis may be put, it is better
to have a specific diagnosis.  So the current system, as currently
implemented, is not working.  In part, that is a problem of
implementation, not the system itself.  But part of the problem is
that the categorical system often does not provide a category that
really fits the patient, so there is a subjective element to the
assignment to a specific category.

A dimensional system might help, by reducing the degree of subjectivity
involved.  The profile of scores on the various scales would be
used to assign the specific category.  If a standardized method
for rating the various scales is used, it could improve some technical
aspects of the diagnostic process, such as the interrater reliability.

Several dimensional models have been proposed.  The main one under
study appears to be the Five-Factor Model.  This has the
unfortunate acronym of FFM.  I say it is unfortunate because I
can’t help but think “flying fructose monster” when I see “FFM.” 
But that is just me.   The DSM sometimes is referred to,
whimsically, as “the psychiatrist’s bible,” so an association to a
whimsical deity is inevitable.

Personally, I have my doubts about the proposed change.  Although
it has potential to be a very helpful innovation, the amount of
research that would have to be done to validate this approach is very
large.  In my view, it would take many years to do an adequate
job.  Implementing such a fundamental change without adequate
research would run the risk of making matters more confusing, not less.

To be continued.


  1. #1 WotWot
    April 7, 2009

    “personality disorder not otherwise specified (PDNOS).”

    NOS is a bullshit pseudo-diagnosis whose real function is to allow a ‘formal diagnosis’ to be made despite a lack of clear diagnosis. If a clinician cannot make an unambiguous diagnosis, then they should not be allowed to sneak a dodgy one in through the back door using the NOS rubric. There are already more than enough problems with psychiatric diagnosis, without adding NOS into the mix.

    If I could make only one change to the DSM, it would be to remove all NOS categories. They are useless, dangerous (particularly in encouraging over diagnosis and mistreatment), and certainly dishonest. If the clinician does not know what the problem is, they should just say so. False labels are worse than no labels at all. The pretence of understanding is of no help to anybody.

    Pardon my bluntness, but this NOS nonsense has gone for too long, and unfortunately does not look like abating anytime soon.

  2. #2 dreikin
    April 7, 2009

    First, I’d like to clarify that this scheme would only (mostly) be applicable to personality disorders – the DSM covers a lot of other mental health issues that may/would not benefit from such a change.

    That said, I’m of mixed feelings about using the FFM for such diagnoses. I’m particularly skeptical of the notion that they are biologically based. Yes, to some degree they are, but I’ve not seen anything that would support an assertion about such as made above. Environment appears to have a much more significant effect than genetics in this regard. There have also been issues extending it to other cultures, where it sometimes fits (in whole or part) and sometimes doesn’t (in whole or part), indicating a strong cultural component.

    It’s also based largely on factor analysis – take a bunch of adjectives from the dictionary that describe human behavior patterns, and see what groupings occur. Since language, while not a limiter of thought, does have significant influence (Zuska has a post today on this same subject), which also could account for at least some of the cross-cultural difficulties of the model.

    It’s also not reflective of considerable portions of personality – some traits not readily reflected in it include religiosity, conservativeness, honesty (important regarding several disorders), etc..

    Further, the factors are not fully independent, and show correlations between some scores. Neuroticism and extroversion, for example, tend to go in opposite directions.

    Factors have also been shown to drift over time with a person, so while it’s rather consistent over time (with a predictable long-scale drift), it’s not stable.

    Also, the very fact that it IS a sliding scale means we won’t get rid of “arbitrary diagnostic thresholds” – as long as we keep using such labels, anyway. There could be a case made for ‘training’ in order to move the scales in one direction or another, but I somehow doubt that’s going to replace the categorical system.

    So, unlike what the article said, it is NOT (completely)

    biologically based, universal, temporally stable, and [able to] avoid problems with the DSM-IV axis II categories including high comorbidity and arbitrary diagnostic thresholds.

  3. #3 dreikin
    April 7, 2009

    Btw, this has actually been an idea floating around for a while – here is an APA-published book from 1993 about personality disorders (using the DSM-III-R) and the FFM. Also, as far as I recall, there is research supporting correlations between certain FFM score profiles and personality disorders, so my rant above wasn’t meant to be dismissive of the idea – just indicative of why that quote was not so solid as it appeared.