See related article by Zimmerman, et al, and commentary by First.
Dimensional Diagnosis and DSM-5
DSM-5 is putting great store in dimensionalization. This is a way of assessing psychopathology that assumes that quantitative measures take precedence over qualitative categorization. In this view, severity scores are a more valid way of diagnosing mental disorders than neo-Kraepelinian disease categories. The preference for dimensions fits the DSM-5 agenda to explain psychopathology within spectra, based on a neuroscience model,1 as advocated by the National Institute of Mental Health.2,3 Dimensions are also consistent with observations that the boundaries of mental disorders are fluid and fuzzy, both with each other and with subclinical phenomena or normal variations.4 However, not everyone stops to consider exactly what is being dimensionalized.
Scoring in psychiatry is not like the staging of tumors, based on imaging and pathological findings. Rather, it is rooted entirely in clinical observation and/or self-report data, ie, signs and symptoms. But observable phenomena can provide only indirect clues to underlying endophenotypes. Until psychiatry identifies biological markers for disease, the use of these kinds of dimensions can only be a rough-and-ready expedient.
Personality disorders present a special challenge for diagnosis. These are complex diagnostic constructs whose causes remain poorly understood. While some patients fit neatly into diagnostic prototypes, most do not. This problem has led trait psychologists (and their supporters within psychiatry) to propose a different system. A dimensional point of view has been very well represented on the DSM-5 Personality and Personality Disorders Work Group. That is no accident, since the leaders of the Task Force have wanted to use personality disorder as a "poster child" for dimensionalization, with the ultimate goal of making all diagnoses in psychiatry quantitative.
Enthusiasts for an idea tend to ignore problems and obstacles. Above and beyond the fact that dimensions do not measure endophenotypes, validating scales is a complex and formidable task. DSM‘s dimensional measures generally lack validity and have not been subject to testing procedures long standard in psychology. Trait psychologists who develop instruments to assess personality dimensions spend years determining whether instruments have acceptable psychometrics. For example, while a 5-point Likert scale has long been a standard way of making ratings quantitative, one must take pains to determine whether measurements are truly continuous or simply ordinal. Unless true dimensionality is established, Likert-style scaling is not necessarily superior to a choice of "yes, no, or maybe."
Severity is a strong predictor of outcome in personality disorder,5 as it is in the rest of psychiatry. That is why proposals for DSM-5,6 as well as for ICD–11,7 recommend using Likert-type scales to measure severity. There is some merit to the idea. But we do not know the psychometrics of such ratings, particularly if made by clinicians with variable training and skill.
Zimmerman et al8 point out that these complex procedures may not even be necessary if, as in DSM-IV, one can score traits as
absent, subclinical, or clinically diagnosable. While that method of scoring is a rough procedure, so are the procedures under consideration for DSM-5. By and large, dimensional scoring makes the most sense for subclinical problems, and less sense for diagnosable and severe personality disorders. Most trait measures have been validated in community populations, not in clinical settings. It
remains to be seen whether describing patients as having a severe personality disorder, and then describing their traits, can do justice to the sickest patients we see.
Dimensions could be most helpful with the problem of personality disorder, not otherwise specified (PD-NOS), which DSM-5 plans to call personality disorder trait-specified.9 These terms describe patients who meet overall criteria for a personality disorder but do not fit into any of the well-established categories. Zimmerman et al10 had found PD-NOS to be the most frequent diagnosis in clinical practice when systematic methods are applied, even if clinicians did not necessarily recognize the clinical picture. It would be advantageous to develop trait profiles for such cases, but severity ratings could suffer from the same problems as those described above.
In summary, the findings of the present study can be viewed as showing that the emperor has no (or few) clothes. Complex scoring of the severity of psychopathology looks like a scientific procedure. But it may not be much better than what we already have.
Author affiliations: Department of Psychiatry, McGill University and SMBD Jewish General Hospital, Montreal, Quebec, Canada.
Potential conflicts of interest: None reported.
Funding/support: None reported.
REFERENCES
1. Regier DA, Narrow WE, Kuhl EA, et al. The conceptual development
of DSM-V. Am J Psychiatry. 2009;166(6):645-650. PubMed doi:10.1176/appi.ajp.2009.09020279
2. Insel TR. Translating scientific opportunity into public health impact: a strategic plan for research on mental illness. Arch Gen Psychiatry. 2009;66((2):128-133. PubMed doi:10.1001/archgenpsychiatry.2008.540
3. Sanislow CA, Pine DS, Quinn KJ, et al. Developing constructs for psychopathology research: research domain criteria. J Abnorm Psychol. 2010;119(4):631-639. PubMed doi:10.1037/a0020909
4. Kendler KS, First MB. Alternative futures for the DSM revision process: iteration v paradigm shift. Br J Psychiatry. 2010;197(4):263-265. PubMed doi:10.1192/bjp.bp.109.076794
5. Hopwood CJ, Malone JC, Ansell EB, et al. Personality assessment in
DSM-5: empirical support for rating severity, style, and traits. J Pers Disord. 2011;25(3):305-320. PubMed doi:10.1521/pedi.2011.25.3.305
6. Livesley WJ. An empirically-based classification of personality disorder.
J Pers Disord. 2011;25(3):397-420. PubMed doi:10.1521/pedi.2011.25.3.397
7. Crawford MJ, Koldobsky N, Mulder R, et al. Classifying personality
disorder according to severity. J Pers Disord. 2011;25(3):321-330. PubMed doi:10.1521/pedi.2011.25.3.321
8. Zimmerman M, Chelminski I, Young D, et al. Does DSM-IV already capture the dimensional nature of personality disorders? J Clin Psychiatry. 2011;72(10):1333-1339.
9. T 06 Personality Disorder Trait Specified. DSM-5 Web site. http://www.dsm5.org/ProposedRevision/Pages/proposedrevision.aspx?rid=23. Updated June 21, 2011. Accessed July 21, 2011.
10. Zimmerman M, Rothschild L, Chelminski I. The prevalence of
DSM-IV personality disorders in psychiatric outpatients. Am J Psychiatry. 2005;162(10):1911-1918. PubMed doi:10.1176/appi.ajp.162.10.1911
