Date(s) - 28/05/2013
2:00 pm - 3:00 pm
SGDP seminar room
Undesired drug effect identification in Danish electronic patient records
Robert Eriksson (Center for Biological Sequence Analysis, Department of Systems Biology, Technical University of Denmark)
Drugs have tremendous potential to cure and relieve disease, but the risk of unintended and undesirable effects is always present. Several of these effects are never identified in the clinical trials prior to authorization, proving drug safety surveillance must continue during marketing. Medical records could be used to identify drug related adverse events and as healthcare providers increasingly store data in electronic patient records, such resources are becoming available to high-throughput text processing.
In the talk the methods used to extract undesired drug effects in Danish clinical narratives will be presented. This includes the developed dictionaries and tools. Additionally, detected drug-adverse event correlations will be presented and discussed. These correlations between drugs and identified adverse events span from trivial to serious potentially fatal effects. We argue the analysis of this largely unexploited data source could be used for both signal detection and accurate analysis of adverse events in the patient population.