Accelerating Pharmacovigilance using Large Language Models

Pharmacovigilance is the practice of monitoring, assessing, and preventing adverse effects or any other drug-related problems. Pharmacovigilance ensures the post-market safety of pharmaceuticals and plays a crucial role in public health by enhancing drug safety. This discipline involves collecting, analyzing, and reporting data on adverse events, allowing for informed regulatory decisions. Manual systems face challenges in handling data volume, potentially leading to oversight and delays. Automation with advanced technologies can be a practical solution to mitigate these challenges and ensure efficient data management. In this talk, we explain the potential application of Large Language Models (LLMs) in pharmacovigilance. We begin with an overview of the process, covering all the stages of the lifecycle. We emphasize the pivotal process of scrutinizing documents for relevant adverse effects and elucidate the measures that enhance their effectiveness. We delineate our strategy for generating informative summaries, with a specific emphasis on adverse effects and their antecedent occurrences. We emphasize the imperative requirement for factual accuracy validation via the implementation of a fact-checking mechanism. We demonstrate the framework's efficacy with a focus on output fidelity and summary informativeness. We provide quantifiable evidence of the benefits of our method, advocating for the adoption of our framework in pharmacovigilance, and conclude by addressing potential refinements.