Use of BPRS-A percent change scores to identify significant clinical improvement: accuracy of treatment response classification in acute psychiatric inpatients.

Use of Brief Psychiatric Rating Scale [Overall J.E., Gorham D.R., 1988. The Brief Psychiatric Rating Scale (BPRS): recent developments in ascertainment and scaling. Psychopharmacology Bulletin 24, 97-99] percent change scores (PCSs) to measure treatment effects may be problematic because two different item-weighting systems (0-6 and 1-7) have been employed to represent the seven rating options and PCSs have demonstrated sensitivity to the item-weighting system used. This study compared the ability of a range of BPRS total scale PCS categories generated by both item-weighting procedures to predict estimates of clinical improvement in a large (N = 1415) heterogeneous acute sample of adult psychiatric inpatients. Results revealed significant differences between the two scaling systems in the proportion of patients classified into categories of PCS symptom improvement. Additional analysis suggested different optimal predictive PCS classifications for each item-weighting system: > 19% for 1-7 and > 39% for 0-6. Guidelines for BPRS publications are presented to facilitate study interpretation and replication. In light of their demonstrated limitations, it is suggested that the use of BPRS PCSs to measure treatment effects be reconsidered.