The effects of MMPI assessment information versus clinical case history information on antisocial personality disorder diagnosis

Jane M Houghtaling, Purdue University

Abstract

Previous results (Houghtaling, Merritt, Witt, Burns & Kost, 1994) found clinicians did not assign antisocial personality disorder diagnoses very accurately when given case history information, especially for female clients. In the present study, both a case history and an MMPI profile of a fictional client, were given to clinicians of both genders. Client gender, assessment condition, and presentation order were manipulated. The purpose of the present study was to determine whether adding assessment information would allow clinicians to use a more statistical approach to boost diagnostic accuracy, and whether it would reduce the inaccuracy associated with female clients. The case history met antisocial diagnostic criteria, and had sub-clinical borderline personality disorder presence. The MMPI was a 4-9 profile, typical of antisocial personality disorder (Greene, 1991). Half of subjects received the case history first, while half received the MMPI first. They then made a tentative diagnosis. Each was then given the supplemental information, and asked to make final diagnoses. Results were analyzed at both stringent categorical-levels (where the full-clinical diagnosis criteria were met), and less-stringent dimensional-levels (where both full-clinical presence and sub-clinical presence constituted the dependent variable). As predicted, clinicians generally were more accurate when they had only the MMPI profile, than with only case history information. However, final diagnoses did not differ between the two conditions; i.e., clinicians were equally accurate after examining both pieces of information. Client gender differences typically were non-significant. A three-way interaction for the dimensional-level, tentative diagnosis analysis, found that female clinicians were more accurate with female clients in the MMPI-first, but not case history-first, condition; male clinicians were more accurate with male than female clients, in the MMPI-first condition. Findings suggest clinicians used theoretical prototypes, sometimes erroneously including male gender as a prototype of antisocial disorder. Implications are: clinicians are able to incorporate probabalistic information; clinicians should use assessment information in formulating diagnoses; and clinicians should be careful to examine information that may not fit their theoretical prototypes. Additionally, both researchers and clinicians would likely benefit incorporating a dimensional approach to diagnosis.

Degree

Ph.D.

Advisors

Merritt, Purdue University.

Subject Area

Psychotherapy|Personality|Psychological tests

COinS