| Session Title: Application of Computerized Adaptive Testing in Clinical Assessment |
| Demonstration Session 869 to be held in International Room on Saturday, November 10, 3:30 PM to 5:00 PM |
| Sponsored by the Quantitative Methods: Theory and Design TIG |
| Chair(s): |
| Kendon Conrad, University of Illinois, Chicago, kjconrad@uic.edu |
| Presenter(s): |
| Barth Riley, University of Illinois, Chicago, bbriley@chestnut.org |
| Michael Dennis, Chestnut Health Systems, mdennis@chestnut.org |
| Abstract: The demand for increasingly complex clinical assessments that cover more areas and constructs mean that such assessments can take 1-2 hours, resulting in respondent burden and requiring considerable staff resources. Computerized adaptive testing (CAT) is a promising approach for improving assessment efficiency, but faces several challenges when applied to clinical assessment. Following an overview of CAT, we will discuss three important issues regarding CAT in clinical assessment. First, we will demonstrate various approaches to estimating measures on multiple clinical domains during a single CAT administration of a clinical assessment battery. Second, we will discuss how precision can be varied across the measurement continuum in order to maximize assessment efficiency and accuracy in clinical decision making. Third, we will discuss the use of fit statistics in the context of CAT to identify persons who present with atypical patterns of symptoms (e.g., persons who endorse suicidal ideation but are not depressed). |