Refinement of phenomenological classification in clinical psychopharmacology research.

A large clinical psychopharmacology data bank consisting of results from several hundred clinical drug trials was used to refine phenomenological classification concepts and methodology. Brief Psychiatric Rating Scale (BPRS) profiles for 2623 psychiatric patients from that data bank were analyzed to define prototypical patterns that are maximally representative of homogeneous subgroup within the drug treatment population. Prototype profiles and classification functions that can be used to accomplish similar classifications of future research subjects are presented. Placebo response rates and active drug response rates for subjects in the large data pool are compared across the eight phenomenological types. It is suggested that anxious and hostile subtypes of depressive disorder may be poor subjects in clinical drug trials because of the excessive placebo response rate. In contrast, withdrawn-disorganized thinking disturbance types tend to be poor subjects for trials of neuroleptic drugs because they respond poorly to both active drug and placebo.