Poster Session

Start Date

20-5-2016 12:30 PM

Document Type

Poster Abstract

Description

Analysis of postmarket drug surveillance reports is imperative to ensure drug safety and effectiveness. FAERS (FDA Adverse Event Reporting System) is a surveillance system that monitors Adverse Events (AEs) from drugs and biologic products. The AEs are reported through MedWatch voluntary reports (initiated from patients and healthcare providers) and mandatory reports (initiated from manufacturers). Much of the information in the voluntary AE reports is narratives or unstructured text. The increasing volume of individual reports, estimated at more than one million per year, poses a challenge for the staff to review large volume of narratives for drug clinical review. We are developing a computational approach using Natural Language Processing and UMLS MetaMap biomedical software to parse the narratives, recognize named-entities in the text and extract consumer/patient and related drug indications and adverse drug reaction information. The goal is to develop a text mining tool that automatically extracts relevant information from the report narratives which can be stored in pre-defined data fields in the FAERS database for efficient searching and querying during clinical review process.

Keywords

text mining, drug safety and effectiveness, adverse events

Creative Commons License

Creative Commons Attribution-Noncommercial-Share Alike 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.

 
May 20th, 12:30 PM

Text Mining From Drug Surveillance Report Narratives

Analysis of postmarket drug surveillance reports is imperative to ensure drug safety and effectiveness. FAERS (FDA Adverse Event Reporting System) is a surveillance system that monitors Adverse Events (AEs) from drugs and biologic products. The AEs are reported through MedWatch voluntary reports (initiated from patients and healthcare providers) and mandatory reports (initiated from manufacturers). Much of the information in the voluntary AE reports is narratives or unstructured text. The increasing volume of individual reports, estimated at more than one million per year, poses a challenge for the staff to review large volume of narratives for drug clinical review. We are developing a computational approach using Natural Language Processing and UMLS MetaMap biomedical software to parse the narratives, recognize named-entities in the text and extract consumer/patient and related drug indications and adverse drug reaction information. The goal is to develop a text mining tool that automatically extracts relevant information from the report narratives which can be stored in pre-defined data fields in the FAERS database for efficient searching and querying during clinical review process.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.