Evidence-Based Assessment of Personality Disorders Thomas A. Widiger and Douglas B. Samuel University of Kentucky The purpose of this article is to provide a foundation for the development of evidence-based guidelines for the assessment of personality disorders, focusing in particular on integrated assessment strategies. The general strategy recommended herein is to first administer a self-report inventory to alert oneself to the potential presence of particular maladaptive personality traits followed by a semistructured interview to verify their presence. This strategy is guided by the existing research that suggests particular strengths of self-report inventories and semistructured interviews relative to unstructured clinical interviews. However, the authors also consider research that suggests that further improvements to the existing instruments can be made. The authors emphasize, in particular, a consideration of age of onset, distortions in self-perception and presentation, gender bias, culture and ethnicity, and personality change. The purpose of this article is to provide a foundation for the development of evidence-based guidelines for the assessment of the personality disorders, including those in the American Psychi- atric Association’s Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM–IV; 1994). We focus on integrated as- sessment strategies rather than specific personality disorder instru- ments. We also emphasize issues that are considered to be funda- mental to a valid personality disorder assessment, being informed by what is generally known about personality disorders, their developmental course, and their comorbidity. We also address matters of clinical utility, because an assessment strategy is un- likely to be effective if it is unrealistic to implement or fails to address matters of importance to clinical intervention. Strategies For the purpose of obtaining an accurate assessment of a DSM–IV personality disorder, the general strategy recommended herein is a two-step procedure: (a) Administer a self-report inven- tory to alert oneself to the potential presence of particular mal- adaptive personality traits, and (b) administer a semistructured interview to verify and document their presence (Widiger, 2002). This integrated strategy is guided by the existing evidence that suggests particular strengths of self-report inventories and semi- structured interviews relative to a reliance on unstructured clinical interviews. However, no method is infallible, and there is compel- ling research to suggest ways in which existing self-report inven- tories and semistructured interviews can be improved. We discuss the rationale and empirical support for using semistructured inter- views and self-report inventories and present potential limitations that could be addressed in future research. Semistructured Interviews The preferred method for assessing personality disorders in general clinical practice appears to be an unstructured clinical interview (Watkins, Campbell, Nieberding, & Hallmark, 1995; Westen, 1997), whereas the preferred method in research is the semistructured interview (Rogers, 2001; Segal & Coolidge, 2003; Zimmerman, 2003). Semistructured interviews have a number of advantages over unstructured interviews (Rogers, 2003). Semi- structured interviews ensure and document that a systematic and comprehensive assessment of each personality disorder diagnostic criterion has, in fact, occurred. The administration of a semistruc- tured interview is particularly advantageous in clinical situations in which the credibility or validity of the assessment might be ques- tioned (e.g., forensic or disability evaluations), because the admin- istration of the interview will document that the assessment was reasonably comprehensive, replicable, and objective. Studies have indicated that diagnoses based on unstructured clinical interviews often fail to consider all of the necessary diagnostic criteria (Blashfield & Herkov, 1996; Zimmerman & Mattia, 1999). As Cantor and Genero (1986) demonstrated, “There is a tendency, once having categorized, to exaggerate the similarity among nonidentical stimuli by overlooking within-group variabil- ity, discounting disconfirming evidence, and focusing on stereo- typic examples of the category” (p. 235). Clinicians tend to diag- nose personality disorders hierarchically, failing to assess for additional symptoms once a particular disorder has been identified (Adler, Drake, & Teague, 1990; Blashfield & Flanagan, 1998; Herkov & Blashfield, 1995). The personality disorder that is provided preferential attention may even be based on idiosyncratic interests (Gunderson, 1992; Mellsop, Varghese, Joshua, & Hicks, 1982). Studies also indicate that personality disorder assessments in the absence of structured clinical interviews can be quite unreliable (Mellsop et al., 1982; Spitzer, Forman, & Nee, 1979), and that semistructured interviews increase the likelihood that a reliable and replicable assessment will occur (Farmer, 2000; Rogers, 2001, 2003; Segal & Coolidge, 2003; Wood, Garb, Lilienfeld, & Nez- worski, 2002). Semistructured interviews provide specific, care- fully selected questions for the assessment of each diagnostic Thomas A. Widiger and Douglas B. Samuel, Department of Psychology, University of Kentucky. Correspondence concerning this article should be addressed to Thomas A. Widiger, Department of Psychology, University of Kentucky, Lexing- ton, KY 40506-0044. E-mail: widiger@uky.edu Psychological Assessment Copyright 2005 by the American Psychological Association 2005, Vol. 17, No. 3, 278 –287 1040-3590/05/$12.00 DOI: 10.1037/1040-3590.17.3.278 278