Differentiating Psychotic Patients From Nonpsychotic Patients With the MMPI–2 and Rorschach

Abstract

The goal of this study was to examine the incremental validity and the clinical utility of the Minnesota Multiphasic Personality Inventory–2 (MMPI–2; (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) and Rorschach (Rorschach, 1942) with regard to differential diagnosis in a sample of adult inpatients with a primary psychotic disorder or a primary mood disorder without psychotic features. Diagnostic efficiency statistics have suggested that the Rorschach Perceptual Thinking Index (PTI; Exner, 2000a, 2000b) was better than MMPI–2 scales in discriminating psychotic patients from nonpsychotic patients. We compared the 84% overall correct classification rate (OCC) for the PTI to an OCC of 70% for the MMPI–2 scales. Adding the MMPI–2 scales to the PTI resulted in a decrease in OCC of 1%, whereas adding the PTI to the MMPI–2 resulted in an increase in OCC of 14%. Sensitivity, specificity, positive predictive power, negative predictive power, and kappa were equal or higher with only the PTI in the model.

Acknowledgment

Tam K. Dao is currently a postdoctural fellow at the Michael E. De Bakey VAMC, Houston, Texas. He is supported by the Office of Academic Affiliations, VA Special MIRECC Fellowship Program in Advanced Psychiatry and Psychology, Department of Veterans Affairs.

Notes

The Goldberg Index (1965) was included in subsequent analyses based on the advice of a reviewer. Given that the Rorschach PTI can be considered as a complex index, the inclusion of the Goldberg Index allowed the Rorschach PTI to be evaluated against a comparable MMPI–2 index. A T score greater than 45 on the Goldberg Index suggests a psychotic profile pattern, and a T score score of 44 or below suggests a neurotic pattern.

*p < .01

** p < .001

^a N = 236

^b n = 108

^c n = 128

*p < .05

** p < .001.

*p < .05.

** p < .001.