When there are several reports of adverse reactions to particular drug, causality assessment may lead to signal generation - a notice of the need for increased awareness of a possible safety problem communicated to countries. Initially signal detection was done by qualitative evaluation and expert review but as the number of reports increased this process became exhaustive and less effective, leading to the development of quantitative tools like Bayesian propagation neural network developed by the UMC. Given the high sensitivity of algorithms, serial testing combination with data mining methodologies for signal detection like BCPNN method [2] might increase causality assessment specificity, deserving further investigation.