Il modello di O.F. Kernberg per lo studio dei disturbi di personalità. Validazione delle versioni italiane dell'inventory of personality organization (IPO) e della structured interview of personality organization (STIPO)

Objective. This work aims at assessing the psychometric properties of the Italian versions of the Inventory of Personality Organization (IPO; Lenzenweger et al. 2001), a self-report instrument for the assessment of personality organization, according to Otto Kernberg’s model, and the Structured Interview of Personality Organization (STIPO), the interview-version of the IPO. Study 1 and Study 2 (Chapter 2) evaluated the psychometric properties of the IPO. Methods. 696 university students and people from the community (Study 1; males 37%; mean age=36.51 ys; SD=±14.08; range: 18-74) and 121 psychiatric patients (Study 2; males 43.8%; mean age=37.22 ys; SD=±10.54; range: 18-66) completed a set of questionnaires including the IPO, the Severity Indices of Personality Problems (SIPP-118; Verheul et al., 2008), the Borderline Personality Disorder Checklist (Giesen-Bloo et al., unpublished manuscript), the Response Evaluation Measure 71 (Steiner et al., 2001), and the Symptom Checklist 90-R (Derogatis, 1977). Results. Confirmatory factor-analyses conveyed results similar to those obtained by Ellison and Levy (2011) and supporting a 4-factor solution (instability of sense of self and others, instability of goals, instability of behaviors, and psychosis). The factor structure proved to be invariant across samples (non clinical and clinical) and gender. The four IPO subscales showed good levels of internal coherence (Cronbach’s alphas between .72 and .91), and good test-retest reliability in the non clinical sample (time frame: one month; ICC ranging between .80 and .96). Patterns of associations with the criterion measures, assessed through path analysis, emerged in line with theoretical expectations. The first factor, Instability of sense of self and others, showed associations with measures of identity integration and defensive functioning, as well as associations with borderline symptoms and psychopathological distress. The second factor, Instability of goals, showed associations with coherent measures of the capacity to set realistic goals and to achieve them. The third factor, Psychosis, showed the expected associations with paranoid/dissociative symptomatology only in the clinical sample. Finally, the fourth factor, Instability of behaviors, was found to be associated with personality functions and symptom clusters connected with difficulties in the regulation of anger, impulses, and aggression. The four factors showed the expected criterion relations: all the dimensions, in fact, discriminate between clinical and non clinical subjects, whereas only Instability of sense of self and others and Instabilty of goals discriminate between borderline patients and non borderline patients. The study presented in Chapter 3 evaluated the psychometric properties of the STIPO. Methods 30 university students and people from the community (males 23%; mean age=40.97 ys; SD=±14.35; range: 22-63) and 49 psychiatric patients (males 46.8%; mean age=36.60 ys; SD=±9.45; range: 20-53) were interviewed through the STIPO and completed a set of questionnaires including the IPO, REM 71 (Steiner et al., 2001), and the SIPP-118 (Verheul et al., 2008); psychiatric patients were also interviewed through the Structured Clinical Interview for Axis II Disorders (SCID II; First et al., 2001). Results Confirmatory factor-analyses confirmed the theoretical 7-factor dimensionality of the interview (Identity, Object relations, Defenses, Coping, Aggression, Moral values, and Reality testing). The STIPO subscales showed good levels of internal coherence (Cronbach’s alphas between .78 and .91), and good inter-rater reliability (ICC ranging between .82 and .97). For what concerns the three primary scales of the STIPO (Identity, Defenses, and Reality testing), the patterns of associations with the criterion measures, assessed through path analysis and regressions, were in line with theoretical expectations. The Identity scale showed associations with measures of the stability of self image and the capacity of pursuing goals. The Defenses scale resulted to be associated with the external measure of primitive defenses and with the domains of the IPO concerning the Instability of the sense of self and others and the Instability of goals, as well as with measures of lack of self-control and emotional instability. Finally, the Reality testing scale showed coherent associations with the Psychosis factor of the IPO. The three primary scales of the STIPO also showed the expected criterion relations: all the dimensions, in fact, discriminate between clinical and non clinical subjects, whereas only Identity and Defenses, and not Reality testing, discriminate between borderline patients and non borderline patients. Conclusions. Our results suggest that the Italian versions of the IPO and the STIPO are reliable and valid tools for the assessment of personality organization according to Kernberg’s model. Results are discussed in terms of their implications regarding research and clinical practice, as well as in the context of the new conception of personality disorders proposed for the DSM V.