•  
  •  
 

Journal of Undergraduate Research

Keywords

change patterns, mentally ill patients, SPMI, brief psychiatric rating scale expanded, BPRS-E

College

Family, Home, and Social Sciences

Department

Psychology

Abstract

This study was conducted to determine differential item sensitivity of The Brief Psychiatric Rating Scale, Expanded (BPRS-E) with an SPMI population. Using item sensitivity information, we tested different factor solutions of the BPRS-E to assess how predictive these are of patient improvement. Although I originally planned to compare BPRS-E scores with scores from the Life Status Questionnaire (LSQ) in this report, I will focus on item sensitivity only. As observed by Vermeersch, Lambert, & Burlingame (2000), while it is important to achieve high levels of reliability when constructing a measure of psychological traits, measures designed to assess patient change need to acknowledge the vital nature of sensitivity to change as a core element. Item sensitivity was tested using the MIXED procedure in SAS (SAS Institute Inc., 1996), (Vermeersch et al., 2000) which is a multilevel, or hierarchical, linear modeling procedure. In essence, this procedure allows for the statistical analysis of time series, repeated measures process data via the generation of an individual slope and y-intercept for each individual subject on the items and total score of the BPRS-E. Such analysis helps determine if change is in the desired direction (i.e., a negative slope indicates patient improvement over time) and whether each individual item’s slope is significantly different from zero (no change). This methodology is not suggested as a means to evaluate patient change using single test items, but rather as a way of selecting items that will enhance the sensitivity to change of the entire measure.

Download

Included in

Psychology Commons

Share

COinS