Toward an empirically based classification of personality pathology.

Abstract

A number of investigations have utilized factor analysis or similar data analytic methods to examine the empirical validity of the Diagnostic and Statistical Manual of Mental Disorders classification system of Axis II personality pathology. This article reviews analyses of the Axis II cluster structure and the latent structure of individual personality disorder criteria. Overall, these studies do not provide sound empirical support for the current personality disorder organization described in the Diagnostic and Statistical Manual of Mental Disorders . They highlight the need for identifying the latent dimensions of personality pathology in order to create a different representation that would more accurately correspond to both a theoretical and functional model of personality disorder. Preliminary research identifying consensus across datasets is summarized. Clinical implications of these findings and future directions for research on personality pathology are discussed.

Comorbidity of personality disorders in anxiety disorders: A meta-analysis of 30 years of research

“…The present findings provide tentative support for this separation as the frequency of cluster C PDs are significantly larger in panic disorder patients with agoraphobia than those not having agoraphobia ( Table 2). The empirical support may however be less valid as it rests on a cluster model of PDs, which has poor validity (Sheets and Craighead, 2007). However, as the clinical severity of panic patients with agoraphobia appears more pronounced than in those without agoraphobia (Furakawa et al, 2009), and as the presence of a comorbid PD is considered highly influential for the natural course and treatment prognosis (Reich, 2003;Telch et al, 2011), the proposed separation converges with the present findings.…”

“…These cluster level differences were only valid for panic disorders and OCD as the other AD studies provided less information about the PD cluster level comorbidity. As the cluster levels will not be retained in the DSM-5 due to weak empirical support (Sheets and Craighead, 2007), this is not relevant to address in future research.…”

The Structure and Correlates of Self-ReportedDSM-5Maladaptive Personality Traits: Findings From Two German-Speaking Samples

“…Regarding the latter issue, there is considerable empirical evidence that the structures of normal and maladaptive personality traits can be jointly and efficiently modeled with four to five higher-order factors (Markon, Krueger, & Watson, 2005;Trull & Durrett, 2005;Widiger & Simonsen, 2005). In contrast, the factor structure of the individual PD criteria in DSM-IV (or ICD-10) is hardly in line with the assumption of 10 distinct higher-order PD categories (Sheets & Craighead, 2007;Wright & Zimmermann, in press). Thus, integrating an empirically supported, hierarchical trait model into the PD sections of DSM-5 and ICD-11 might help improve the psychometric properties of PD diagnoses.…”

The structure of personality pathology: Both general (‘g’) and specific (‘s’) factors?

“…Furthermore, meta-analytic studies concur (O'Connor, 2005;O 'Connor & Dyce, 1998), leading Sheets and Craighead (2007) to conclude in their review that studies examining the structure of PD criteria have generally not supported the DSM organization. A few exceptions to this conclusion have been studies finding evidence for a nine-or 10-factor solution that approximates the DSM PD structure (Black burn, Logan, Renwick, & Donnelly, 2005;Durrett & Westen, 2005;Howard, Huband, Duggan, & Mannion, 2008;Huprich, Schmitt, Richard, Chelminski, & Zimmerman, 2010).…”