Online Program Home
My Program

Abstract Details

Activity Number: 499
Type: Contributed
Date/Time: Wednesday, August 3, 2016 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Learning and Data Science
Abstract #319957
Title: Statistics and Machine Learning in Pharmacovigilance for Signal Detection of Cardiovascular Risks
Author(s): James Chen* and Weizhong Zhao and Wen Zou
Companies: FDA/NCTR and FDA/NCTR and FDA/NCTR
Keywords: Adverse event detection ; disproportionality analysis ; drug-drug association ; Post-marketing Surveillance ; Text mining
Abstract:

Drug-induced cardiovascular adverse events have been a major concern for the Food and Drug Administration (FDA) and pharmaceutical companies. Drug-induced cardio-toxicity is observed in association with classes of drugs, including anticancer, antibiotics, antidepression, antipsychotics, and antidiabetics. FDA Adverse Event Reporting System (FAERS) has been a core pharmacovigilance system to support the FDA post-marketing safety surveillance program for all approved drug and therapeutic products. Current methods focus on identifying high reporting rates in a particular drug and a particular adverse event (AE). However, a single drug and AE combination should not be considered in isolation, but combinations of other AEs with that drug and other combinations of that AE and that drug should be considered. This study utilizes machine learning algorithms to identify sets of drugs that share common profiles of AEs and their relationships, and develops statistical models for signal detection and analysis of two particular reporting rates of interest, such as sex-related differences in drug-induced cardiac risks.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

 
 
Copyright © American Statistical Association