Statistical data mining applied to large databases of spontaneous post-marketing adverse event reports has demonstrated substantial utility in supporting pharmacovigiiance efforts at government health authorities and pharmaceutical manufacturers. The objective of this research is to develop advanced statistical methods to help assess, in the safety data mining context, the contribution of individual drugs to adverse events in situations involving polytherapy. This work addresses an important theoretical and practical problem in pharmacovigilance, and builds on a long-standing collaboration which has led to the development and deployment of applicant's safety data mining software at FDA and in industry. In Phase I, the focus is on the implementation and evaluation of a method utilizing information from three-way (drug-drug-event) combinations to calculate a polytherapy adjustment score for each drug-event pair and thus to provide guidance on whether a high signal score may be due to the presence/influence of other drugs. Phase II involves extensions to the algorithm, full integration of the algorithm into applicant's safety data mining system, and preparation of educational and tutorial materials. The key benefit to public health is an improved ability to interpret signals of potential drug safety problems, and to prioritize efforts to investigate and evaluate these signals.
