Discriminant Analysis (proc Discrim) to Separate a Study Population by Treatment Subgroups in a Clinical Trial with a New Antidepressant

Data were evaluated from a pilot phase 2 clinical trial (N=55) with a new antidepressant and multivariate discriminant analysis (PROC DISCRIM) was used to separate the drug-treated from placebo populations by treatment subgroups. The list of variables for discriminant analysis included the percent of change from baseline for the primary efficacy variable (21-item Hamilton Depression Rating Scale – HAMD-21) for 11 major time points from Day 3 (the first day of evaluation after treatment) to Day 40 (the last day of evaluation at the end of the trial). Separation with discriminant analysis was very effective for any combination of two of three treatment subgroups (from 82.4% to 86.7% of subjects with the correct classification by treatment subgroups) and for the three treatment subgroups together (80% of subjects with correct classification).