Journal of Personality Disorders, 29, 2015, 231
© 2015 The Guilford Press
1
From Department of Psychology, Milan Bicocca University, Milan, Italy (E. S. G., M. L.); Department
of Psychological Sciences, Purdue University, West Lafayette, Indiana (D. B. S.); and Associazione per la
Ricerca in Psicologia Clinica - A.R.P., Milan, Italy (M. L.).
We deeply thank the patients, clinicians, and staff of the Associazione per la Ricerca in Psicologia Clinica
(A.R.P., Milan, Italy) where this study was conducted for the time and knowledge they shared with us. We
also acknowledge all the colleagues with whom we had the opportunity to productively discuss over the
years about the main topics discussed in this work.
Margherita Lang served as a co-editor for the Italian version of the MCMI-III and as such receives royalties
for its use. No funding was received for this research.
Address correspondence to Emanuela S. Gritti, Department of Psychology, Milan Bicocca University, Piazza
dell’Ateneo Nuovo 1, Milan 20126, Italy; E-mail: emanuela.gritti@unimib.it
DIAGNOSTIC AGREEMENT BETWEEN
CLINICIANS AND CLIENTS:
THE CONVERGENT AND DISCRIMINANT
VALIDITY OF THE SWAP-200 AND
MCMI-III PERSONALITY DISORDER SCALES
Emanuela S. Gritti, PhD, Douglas B. Samuel, PhD,
and Margherita Lang, PsyD
A particularly controversial aspect in the feld of personality assessment is
the use of self-report measures, versus clinicians’ evaluations, for diagnosing
personality disorder (PD). No studies have systematically documented
the agreement between these sources for the entire array of DSM-5 PDs
using comprehensive measures and experienced clinicians’ judgments. The
present work flls this gap by indexing the agreement between patients’
self-descriptions and clinicians’ judgments, relying on standardized and
thorough PD instruments. The Shedler-Westen Assessment Procedure–200
(SWAP-200; Westen & Shedler, 1999a, 1999b) and the Millon Clinical
Multiaxial Inventory-III (Millon, Davis, & Millon, 1997) were both
completed in a clinical series of 56 adult outpatients. Analyses highlighted
moderate correlations between the two measures for the 10 DSM-5 PDs
(Mdn = .35). Agreement was highest for psychological features that are
more easily observable by the clinicians. Furthermore, results revealed
problematic discriminant validity between the two instruments.
The diagnosis and assessment of personality disorders (PDs) has traditionally
been a thorny issue that has created complications. Within practice settings,
PD diagnoses are primarily assigned by treating therapists based on their
clinical interactions and unstructured interviews with the patient (Perry, 1992;
Westen, 1997). Nonetheless, the past three decades have seen a proliferation
G4361_231.indd 1 11/23/2015 9:39:10 AM