NCTR Research Highlights
Get e-mail updates when this information changes.Current Highlight from March 8, 2013
Data Mining for Signal Detection of Adverse Event Safety Data
NCTR scientists, in collaboration with the Center for Drug Evaluation and Research (CDER), have developed a data mining process (algorithm) that allows the simultaneous characterization of the similarities and differences among drugs and adverse events in the FDA Adverse Events Reporting System (AERS) database. The algorithm identifies and describes associations between subsets of drugs and corresponding subsets of adverse events. These “biclusters” can then be investigated and applied to databases of other drugs and adverse event safety data to identify unrecognized adverse event associations with drugs, or highlight unusual anomalies. A safety data set consisting of 193 cardiovascular drugs with 8,543 adverse events was analyzed as an illustration. The results of this study have been published in the Journal of Biopharmaceutical Statistics (2013, 23:146-160).
For additional information, please contact James J. Chen, Ph.D., Biostatistics Branch Director, Division of Bioinformatics and Biostatistics, FDA/NCTR.
-
-
No hay comentarios:
Publicar un comentario