Structured interviews allow for a detailed assessment of psychological functioning and the opportunity for trained interviewers to gather follow-up information for the most relevant domains. They also rely less on clients possessing insight as to their functioning—a significant consideration in some contexts—and offer improved reliability over unstructured interviews (see Segal & Coolidge, 2007, for advantages and disadvantages of interviews). The five interviews listed in Table 12.1 assess all of the DSM-IV PDs. They are the Structured Clinical Interview for DSM-IV Axis II (SCID-II; First, Gibbon, Spitzer, Williams, & Benjamin, 1997); Structured Interview for DSM-IV Personality (SIDP-IV; Pfohl, Blum, & Zimmerman, 1997); International Personality Disorders Examination (Loranger, 1999); Personality Disorders Interview-IV (PDI-IV; Widiger, Mangine, Corbitt, Ellis, & Thomas, 1995); and Diagnostic Interview for Personality Disorders-IV (DIPD-IV; Zanarini, Frankenburg, Sickel, & Yong, 1996). These measures range from 30 to 120 min for typical full administration, but most should take 60–90 min. However, a review of the empirical literature reflects that many researchers choose to administer only portions of these instruments for the purpose of targeting a single or subset of DSM-IV PDs. They target a fair amount of overlapping content (i.e., DSM-IV PD space), with individual criteria typically rated on 3- or 4-point scales ranging from absent to of clinical concern. However, each has unique features that users may consider when selecting the most appropriate one to use. For example, the SCID-II and DIPD-IV are organized by DSM-IV disorder, whereas the SIDP-IV and PDI-IV can be administered either by disorder or by behavioral/thematic domain (e.g., interests and activities, work style). To facilitate broad usage, the IPDE has been translated into several languages and can be scored for either DSM-IV or ICD-10 PDs. Of value in some settings, the SCID-II and the IPDE offer questionnaire-based screening instruments. The literature evaluating the psychometric properties of these five interviews is extensive in some cases, including consideration of their utility in one context versus another (e.g., forensic versus clinical self-referral). Overall, they cover the territory of the DSM-IV PDs; it remains to be seen how much revision they will require as we approach DSM-5.

An alternative clinician-rated measure, the Shedler-Westen Assessment Procedure (SWAP; Shedler & Westen, 2007), is a 200-item Q-set. The SWAP was developed in part to further address the issue of client insight. “The ‘standard vocabulary’ of the SWAP allows clinicians to provide in-depth psychological descriptions of patients in a systematic and quantifiable form and ensures that all clinicians attend to the same spectrum of clinical phenomena” (Shedler & Westen, 2007, p. 43). The SWAP generates dimensional scores for the DSM-IV PDs, a set of empirically derived syndromes, and 12 personality factors, and includes narrative case descriptions. This instrument represents a departure from the interviews above.

Questionnaires offer an efficient and cost-effective means of assessing symptoms and there are many instruments designed to address DSM PDs. Table 12.1 lists six common instruments: Personality Assessment Inventory (PAI; Morey, 2007); Dimensional Assessment of Personality Pathology Basic Questionnaire (DAPP-BQ; Livesley & Jackson, 2009), Millon Clinical Multiaxial Inventory (MCMI-III; Millon, Davis, Millon, & Grossman, 2009), NEO Personality Inventory Revised (NEO-PI-R; Costa & McCrae, 1992); Personality Disorder Questionnaire (PDQ-4; Hyler, 1997); and Schedule for Nonadaptive and Adaptive Personality (SNAP-2; Clark, Simms, Wu, & Casillas, in press). More so than the interviews, these questionnaires reflect substantial variability in content. That is, whereas most ­contain items that can be mapped onto DSM PD criteria—even if such assessment was not the major aim of the instrument—the personality space tends to be broader. In fact, all of the Table 12.1 instruments include content and scales beyond the formal PDs. For example, the PAI includes 18 scales addressing clinical content, treatment, and interpersonal functioning; the DAPP-BQ also assesses 18 traits relevant to personality pathology; beyond the DSM PDs, the MCMI-III assesses 11 clinical syndromes and 42 facets. The SNAP-2 includes 15 trait and temperament scales, in addition to the DSM-IV PD scales which themselves can be scored in three ways: dimensionally, by presence of individual criteria, and for categorical diagnosis. Noted previously, the NEO-PI-R is tied to the FFM and has been used increasingly to assess content relevant to personality pathology. Of note, several of these measures offer the benefit of separate validity scales, valuable for detecting problematic response styles such as random or inconsistent responding, true or false tendency, or symptom exaggeration/minimization. Some include separate forms for the report of informed others, such as parents or spouses. In combination with clinical assessment, these questionnaires can be valuable in the diagnosis and conceptualization process.

As noted, several trait models and associated questionnaires appear in the contemporary ­literature. It is important to restate here that these instruments converge, but they are not redundant—each possesses unique information. For example, Bagby, Marshall, and Georgiades (2005) compared three trait instruments—one representing 5-, 7-, and 18-factor models—to predict interview-based DSM-IV PD symptoms. All three instruments significantly explained PD variance. However, the 18-factor model contributed unique prediction beyond the 5- and 7-factor models for all ten PDs; the 5-factor model contributed unique prediction beyond the 7-factor model for all ten PDs; the 7-factor model contributed unique prediction beyond the 5-factor model for eight of the PDs and beyond the 18-factor model for five of the PDs. Other studies have drawn similar conclusions; from such work it becomes clear that the PD space is broad and there may not yet be a single measure that comprehensively incorporates all of it.