Computerized Adaptive Assessment of Personality Disorder: Introducing the CAT–PD Project

Abstract

Assessment of personality disorders (PD) has been hindered by reliance on the problematic categorical model embodied in the most recent Diagnostic and Statistical Model of Mental Disorders (DSM), lack of consensus among alternative dimensional models, and inefficient measurement methods. This article describes the rationale for and early results from a multiyear study funded by the National Institute of Mental Health that was designed to develop an integrative and comprehensive model and efficient measure of PD trait dimensions. To accomplish these goals, we are in the midst of a 5-phase project to develop and validate the model and measure. The results of Phase 1 of the project—which was focused on developing the PD traits to be assessed and the initial item pool—resulted in a candidate list of 59 PD traits and an initial item pool of 2,589 items. Data collection and structural analyses in community and patient samples will inform the ultimate structure of the measure, and computerized adaptive testing will permit efficient measurement of the resultant traits. The resultant Computerized Adaptive Test of Personality Disorder (CAT–PD) will be well positioned as a measure of the proposed DSM–5 PD traits. Implications for both applied and basic personality research are discussed.

Acknowledgments

We thank William Calabrese, Monica Rudick, Wern How Yam, Tom Yufik, and Kerry Zelazny for their help in coordinating the expert review study and the development of the initial item pool. In addition, we thank the experts who provided feedback on the candidate traits as well as the research assistants who sorted items into candidate trait bins. Finally, we thank Lee Anna Clark for her consultation during the early stages of the project.

The studies presented in this article were supported by a research grant to the L. J. Simms from the National Institute of Mental Health (No. R01MH080086).