Use of the TAT in the assessment of DSM-IV cluster B personality disorders

Citation
Sj. Ackerman et al., Use of the TAT in the assessment of DSM-IV cluster B personality disorders, J PERS ASSE, 73(3), 1999, pp. 422-448
Citations number
91
Categorie Soggetti
Psycology
Journal title
JOURNAL OF PERSONALITY ASSESSMENT
ISSN journal
00223891 → ACNP
Volume
73
Issue
3
Year of publication
1999
Pages
422 - 448
Database
ISI
SICI code
0022-3891(199912)73:3<422:UOTTIT>2.0.ZU;2-N
Abstract
The Social Cognition and Object Relations scale (SCORS), developed by Weste n, Lohr, Silk, Kerber, and Goodrich (1985), is a diagnostic instrument used to assess an array of psychological functioning by using clinical narrativ es such as the Thematic Apperception Test (TAT; Murray, 1943) stories. This study investigated the utility of the SCORS to differentiate between Diagn ostic and Statistical Manual of Mental Disorders (4th ed. [DSM-IV]; America n Psychiatric Association, 1994) antisocial personality disorder (ANPD), bo rderline personality disorder (BPD), narcissistic personality disorder (NPD ), and Cluster C personality disorder (CPD). A sample of 58 patients was se parated into four groups: ANPD (n = 9), BPD (n = 21; 18 with a primary BPD diagnosis and 3 with prominent borderline traits who met 4 of the 5 DSM-IV criteria necessary for a BPD diagnosis), NPD (n = 16; 8 with a primary NPD diagnosis and 8 with prominent narcissistic traits who met 4 of the 5 DSM-I V criteria necessary for a NPD diagnosis), and CPD (n = 12). These groups w ere then compared on the 8 SCORS variables by using 5 TAT cards (1, 2, 3BM, 4, and 13MF). Spearman-Brown correction for 2-way mixed effects model of r eliability for the 8 SCORS variables ranged from .70 to .95. The results of categorical and dimensional analyses indicate that (a) the SCORS variables can be used to differentiate ANPD, BPD, and NPD; (b) the BPD group scored significantly lower (greater maladjustment) than did the CPD group on certa in variables; (c) the BPD group scored significantly lower (greater maladju stment) than did the NPD group on all 8 SCORS variables; (d) the ANPD group scored significantly lower than did the NPD group on certain variables; (e ) certain variables were found to be empirically related to the total numbe r of DSM-IV ANPD, BPD, and NPD criteria; and (f) certain variables were fou nd to be empirically related to Minnesota Multiphasic Personality Inventory -2 (MMPI-2; Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) personal ity disorder scales. The results of this study are discussed in terms of cl inical utility, conceptual, and theoretical implications.