Text Mining for Adverse Drug Events: the Promise, Challenges, and State of the Art

Abstract

Text mining is the computational process of extracting meaningful information from large amounts of unstructured text. Text mining is emerging as a tool to leverage underutilized data sources that can improve pharmacovigilance, including the objective of adverse drug event detection and assessment. This article provides an overview of recent advances in pharmacovigilance driven by the application of text mining, and discusses several data sources—such as biomedical literature, clinical narratives, product labeling, social media, and Web search logs—that are amenable to text-mining for pharmacovigilance. Given the state of the art, it appears text mining can be applied to extract useful ADE-related information from multiple textual sources. Nonetheless, further research is required to address remaining technical challenges associated with the text mining methodologies, and to conclusively determine the relative contribution of each textual source to improving pharmacovigilance.

Identification of Risk Factors in Clinical Texts through Association Rules

“…The main limitation was the lack of a systematic evaluation of the developed text mining system. In (Harpaz et al, 2014) the authors state that TM is sufficiently mature to be applied for the extraction of useful information concerning ADEs from multiple textual sources. Currently such information is collected by manual expert analysis of clinical trial notes and spontaneous reports, and the review of biomedical literature; but progress depends on a comprehensive approach that examines a diverse set of potentially complementing data sources including EHRs.…”

Web-Based Signal Detection Using Medical Forums Data in France: Comparative Analysis (Preprint)

“…Recent projects or collaboration aimed at investigating the quality of social media data as well as to investigate the most performant method to do web-based signal detection. The use of webbased data (such as, query logs and social media) is emerging among regulators (FDA and EMA), industry and academia [5,6,[8][9][10][11]. As an example, a public private partnership between the European Commission and European Federation of Pharmaceutical Industries and Associations, called WEB-RADR -Recognising Adverse Drug Reactions has been launched in 2014.…”

Identifying & Classifying Potential Post-Market Drug ADEs from Social Media Blogs: A Research Proposal for Detecting ADEs over Users Conversation Networks

“…Because of the costs associated with the post-marketing ADEs caused by drugs, and the large volume of user posted information available in social media, there is a strong motivation for systems that can automatically monitor social media sites and generate signals when adverse reactions frequently occur for specific drugs [Patki 2014]. The recent advances in the data processing capabilities of computers, machine learning and NLP research presented an excellent possibility of utilizing this massive data source over the social media for a variety of purposes, including ADE monitoring [Harpaz 2014]. Drug regulators like FDA 1 also started to use social media posts for potential adverse drug event signals especially for post-marketing new drugs.…”