DIFFERENT MODES OF DATA-PROCESSING AND STATISTICAL TESTING APPLIED TOTHE SAME SET OF PHARMACO-EEG RECORDINGS - EFFECTS ON THE EVALUATION OF A SELECTIVE AND THE REVERSIBLE MAO-A-INHIBITOR (BROFAROMINE)

The comparison of two different modes of data processing and two diffe rent approaches to statistical testing both applied to the same set of EEG recordings was the main objective of this pharmacological study. Brofaromine (CGP 11 305 A), a new selective and reversible monoamine o xidase type A inhibitor was used as an example for investigating a pot entially antidepressant drug in clinical development. The two modes of pharmaco-EEG (PEEG) data processing differed mainly in the sampling f requency and definition of spectral parameters. Patterns of significan t changes were noted in terms of descriptive data analysis using eithe r a nonparametric Wilcoxon signed-rank test or an ANOVA of transformed data, as suggested by Conover and Iman, These data clearly demonstrat e that slight discrepancies in the results may simply arise from diffe rences in data processing and statistical approach applied, In spite o f these discrepancies, the patten of brofaromine-induced PEEG changes was very similar regardless of the mode of data handling used.