Since the early 1990s, the FDA has used data mining (the practice of examining large databases to generate new information) to gain a better understanding of adverse signals within clinical trial safety data. The agency has also advocated the use of data mining by the pharmaceutical industry when conducting clinical trials. Recently, the FDA has begun adding more sophisticated methods to its data mining activities and has applied these methods to other product-safety-related databases.
In an effort to define its data mining activities, the FDA has issued a white paper, Data Mining at FDA (Nov 2015). It provides an overview of its past and present data mining methods, the advantages/challenges of data mining and future directions for data mining at the FDA.
In relation to past and current data mining methods, the paper describes the FDA’s Center for Drug Evaluation and Research (CDER) application of software solutions -- including Oracle Health Sciences Empirica Study -- to analyze clinical trial data for new drug applications or supplemental applications. The paper also outlines how the FDA uses Oracle Health Sciences Empirica Signal for the routine mining of data related to drugs, foods, cosmetics, and dietary supplements.
In regard to future directions for data mining, the FDA, using Oracle Health Sciences Empirica Signal and Study, is in the early stages of developing an automated data mining capability. When completed, this capability would nearly eliminate the need for any additional labor resources.
Dr. Sameer Thapar, Director, Global Pharmacovigilance (Oracle Health Sciences Consulting), and Assistant Professor, Drug Safety and Pharmacovigilance (Rutgers University), has studied this paper and offered highlights in a new Oracle Brief. Read it here.
Read the FDA white paper Data Mining at FDA
Learn more about Oracle's solutions data mining -- at the FDA and for your clinical trials -- Booth #1125 DIA16.