Problems with the Current Diagnostic System Continued

2. Relative symptom importance and overly broad symptom descriptions:

In addition to being categorical, the diagnostic approach used in DSM IV-TR is also polythetic. This term refers to the fact that in order to be diagnosed with a personality disorder, a person must meet a certain number of symptom criteria from a defined criteria set. For instance, there are nine symptoms (criteria) that define Borderline Personality Disorder. In order to be diagnosed as having Borderline Personality Disorder, a person is required to meet five out of those nine symptom criteria. But, there are several pitfalls to this approach.

The first problem has to do with the fact that a polythetic system presupposes all diagnostic symptoms carry the same weight, and are considered equally important when making a diagnostic determination. It might be reasonable to assume that the equality among symptoms makes the diagnostic process somewhat simpler, but this assumption may not be warranted. For instance, two of the possible nine criteria for the Borderline Personality Disorder include "suicidal and self-harming behavior" and "chronic feelings of emptiness." Whether these two criteria should be considered equally important is open to debate. However, today, many practicing clinicians would consider evidence of self-harming behavior to be a vastly more significant indicator of Borderline Personality Disorder than reported feelings of emptiness. It is true that reported feelings of emptiness are indeed an important symptom, but this symptom can be associated with several other diagnoses as well. Thus, the assumption that equality among symptoms makes for diagnostic simplicity is perhaps an inaccurate one.

Another problem with the polythetic approach is the issue of "high heterogeneity." This term refers to the fact that there is a great deal of variation among people with the same disorder. To illustrate this concept, let's examine the diagnostic requirements for the Borderline Personality Disorder. The Borderline Personality diagnosis requires a person must meet at least five out of nine possible symptom criteria. This means that there are more than 150 possible combinations of five criteria that can qualify a person for a diagnosis of Borderline Personality Disorder. In other words, 150 people with the same diagnosis of Borderline Personality would each have a different set of symptoms. Stated differently, this also means that any two people who are diagnosed with Borderline Personality Disorder need only have one single symptom in common! High heterogeneity is considered problematic because when people with the same disorder do not appear to share similar symptoms, the diagnostic label becomes less meaningful and less useful to clinicians, researchers, and patients alike. As a result, diagnostic labels are seen as unreliable indicators that a particular underlying problem is actually present in the individual bearing that diagnosis.

Another problem with high heterogeneity is a problem that is commonly observed in clinical practice. This problem occurs when a patient does not meet enough criteria for any one, single personality disorder diagnosis to merit that diagnosis being applied, but does meet some criteria within several different personality disorder diagnoses, usually within the same cluster . In this case, it is common to diagnose "Personality Disorder Not Otherwise Specified" (NOS), which is a catch-all diagnostic category used to indicate that something of clinical relevance is present, but nothing that (yet) cleanly fits into established categories. The use of the Personality Disorder NOS diagnosis is extremely commonplace (Verheul and Widiger, 2004), and yet unsatisfying due to its high degree of ambiguity, and lack of precision.

3. The co-occurrence of Axis II Personality Disorders with other disorders

Another diagnostic difficulty is that we now know that people frequently meet criteria for more than one personality disorder at a time. When two or more disorders occur simultaneously, clinicians and researchers call this "co-occurrence." Research has shown that there is a tendency for personality disorders of the same cluster to co-occur (Skodol, 2005). While this finding lends support to the validity of the personality disorder clusters,  it calls into question whether individual personality disorder diagnoses represent discrete entities.

This problem of high co-occurrence has also been observed in conjunction with Axis II personality disorders and Axis I disorders. This finding raises the issue of whether it is sensible and correct to separate personality disorders on Axis II, from the clinical disorders described on Axis I (see also here). Experts have proposed several different explanations for these findings. One explanation is that one disorder (e.g., an Axis II disorder like Borderline Personality Disorder) may precede the other (an Axis I disorder like Major Depression) and increases the risk of developing this second disorder. Another explanation is it may be possible that both disorders are best understood as different manifestations of the same underlying issue, with one disorder representing a milder version of the other. Ongoing research will attempt to make further sense of these and similar observations. However, there is already ample evidence available to suggest that the current DSM personality diagnosis diagnostic system is not the most realistic, nor the most accurate diagnostic system possible. For a more thorough discussion of this issue of co-occurrence please go to this section.

The multi-axial diagnostic system has been controversial ever since its introduction (DSM-III, 1980). The American Psychiatric Association is presently working on the next revision of DSM, DSM-V. There are rumors circulating in the professional community that some disorders may be moved from Axis II to Axis I in that release, or even that the distinction between Axis I and II disorders will be done away with entirely. However, this is all conjecture, as the final structure of the DSM-V has not yet been released at the time of this writing¹, and thus remains to be seen.

1 This document was released for publication on http://www.mentalhelp.net in January 2011.