The diagnosis of personality disorders (PDs) remains controversial, and the primary aim of our original application was to investigate the operating characteristics, especially the reliability and validity, of two alternative methods for diagnosing personality disorders: structured interviews and self-report measures.
The first aim of this competing renewal application is to continue to enlarge our sample and to complete the work described in the previous version (as recommended by the Committee).
The second aim i s to apply novel quantitative approaches to the data archive we have collected. our primary goals for data analysis are; (1) To develop more efficient and valid """"""""stage I"""""""" screening tools for personality disorder, using self-report, clinical, and demographic data analyzed with methods derived from item response theory and signal detection theory; (2) To continue to improve our understanding of the structured interviews better used as """"""""stage II"""""""" diagnostic instruments. These efforts will include examination of the concordance among patient, significant other, and clinical consensus data using designs developed from generalizability theory; analyses of predictive validity, both in terms of clinical outcomes and treatment utilization and associated costs; and further analyses of the usefulness of the interviews for identifying severe, """"""""high risk"""""""" personality disorders; (3) To try to resolve some questions about the dimensionality of the DSM-III-R nomenclature for personality disorder with techniques that exploit the greater reliability and power of dimensional versus categorical data (e.g., exploratory and confirmatory factor analysis, covariance structure analysis). Based on our preliminary analyses, we propose a four-factor model for the current DSM- III-R personality dimensions, with factors for aggression (covert and overt), inhibition versus disinhibition, interpersonal ambivalence, and schizotypy; (4) To develop a more differentiated understanding of the influences that are critical in our clinical consensus judgements. We will develop multivariate models to identify the impairments that are most closely linked to our categorical diagnoses and our continuous rating of Axis II severity. The identification of these areas and the traits (e.g., aggression, disinhibition) that may be associated with them will inform more rational development of treatment models, a long-term future goal; and (5) To continue to explore the convergent relationships between descriptive features of personality disorders, impairments in social functioning, and adult attachment styles. This proposal is a revision of competing continuation application 2 R01 MH44672-05.
Showing the most recent 10 out of 11 publications
