A New Framework for Personality Assessment in DSM-5

Unless you’re an academic psychiatrist with a niche interest within the diagnostic nomenclature of psychiatry, chances are you probably haven’t spent much time poking around the website of the DSM-5 Task Force at dsm5.org. In recent months, the publication date for this much-awaited tome has been pushed back to May 2013. In the meantime, the task force leadership and the diagnosis work groups have used the website to post updates, explanations, and rationales for proposed changes from the DSM-IV-TR framework.

The Personality and Personality Disorders Work Group, headed by Dr. Andrew E. Skodol of the Institute for Mental Health Research in Arizona, has received an outsized share of attention in the lay and professional press, including the New York Times (Nov. 29, 2010, page D1). The work group has proposed extensive and far-reaching changes to the diagnostic structure of Axis II, to the surprise of many observers and commentators.

It’s been widely reported that the DSM-5 work group plans to reduce the number of personality disorder diagnoses to six: antisocial, avoidant, borderline, narcissistic, obsessive-compulsive, and schizotypal. (The Times article reported on a proposal – since rescinded – to eliminate narcissistic personality disorder as well.) Eliminated from the DSM-5 would be the histrionic (hysterical), schizoid, and dependent personality disorders.

This change, although major, is less significant than two other modifications proposed for the DSM-5.

In a second key proposal, the work group adds an innovation previously unseen in the DSM but well known to generations of psychiatry residents from introductory psychotherapy textbooks: an assessment of whether the patient is functioning at a neurotic, borderline, or psychotic level of personality organization. In the DSM-5, this assessment is termed the Levels of Personality Functioning scale, and patients are rated from 0 (healthy) to 4 (extreme impairment). My review of the scale suggests that a 0 rating corresponds to mental health, a 1 rating corresponds to a neurotic level of personality organization, a 2 corresponds to a mild borderline level of personality organization, a 3 corresponds to a severely borderline level of personality organization, and a 4 corresponds to a psychotic level of personality organization.

The third major change proposed by the work group has attracted the most criticism. The work group proposed an entirely separate system of personality assessment, unrelated to the syndrome recognition system from DSM-IV-TR, in which clinicians are asked to rate individual patients on each of five "personality trait domains" (negative affectivity, detachment, antagonism, disinhibition, and psychoticism). These major categories each comprise four to nine "trait facets," each of which can also be assessed by the clinician. For example, the trait facets proposed for the negative affectivity domain include mood lability, anxiety, separation anxiety, perseveration, submissiveness, and depressivity. The dsm5.org website includes a 7-page "DSM-5 Clinicians' Personality Trait Rating Form" that details how each of these trait domains and trait facets should be rated.

So to review, the three major changes proposed by the personality work group are the following:

• Fewer personality syndromes.

• Assessment of the degree of impairment due to personality disturbance.

• Trait-based, nonsyndromic assessment of personality traits.

Of these alterations, the second proposal seems to be the most helpful and effective. Including an assessment of the severity of personality disturbance will enhance the clinical utility of the DSM-5 diagnostic system. A busy clinician will gain valuable information from a diagnostic system that indicates whether an individual has been high functioning despite personality impairment, or instead whether an individual has been partly or completely incapacitated by pervasive personality disturbance.

The first proposal, which would reduce the number of personality disorder types available in the DSM-5 system, appears to have been motivated by the relative paucity of contemporary research about the histrionic, schizoid, and dependent personality types.

On one hand, the work group’s efforts to modernize and update personality diagnosis appear well intentioned; the field of psychiatry should not cling to outdated notions of personality structure that have not been scrutinized and validated. On the other hand, the decision to eliminate the three extant personality disorders from the DSM-5 would effectively cleave those syndromes out of future psychiatric research and practice, and would cut off access to the clinical wisdom accumulated over decades of work with the chronic dramatization of the histrionic patient, the inscrutable withdrawal of the schizoid patient, and the neediness of the dependent patient. If a clinically useful clustering of personality traits does not have a label, then for future generations of psychiatrists, it will not exist.

The third proposal is by far the most controversial. Clinicians would be asked to rate individual personality traits and facets with little regard for pattern recognition. Pattern recognition is, of course, the hallmark not only of psychiatric diagnosis, but of medical diagnosis in general. The doctor who fails to recognize the pattern of chest pain radiating down the left arm with sweating and dyspnea will also fail to treat the acute coronary syndrome.