Abstract

PHARMACOVIGILANCE IN THE 21ST CENTURY: METHODS, CHALLENGES, AND THE IMPACT OF REAL-WORLD DATA AND ARTIFICIAL INTELLIGENCE

Satyam Kumar*, Vikas Saini, Manish Bhargav, Rahul Saini, Abhishk Verma, Pawan Kumar Basniwal

ABSTRACT

Pharmacovigilance (PV) — the science and activities for detecting, assessing, understanding, and preventing adverse effects or other drug-related problems — is essential to ensure medicine safety across their lifecycle. Traditional passive surveillance systems (spontaneous reporting) remain foundational, but they are limited by underreporting and bias. The expanding availability of realworld data (RWD), advances in signal detection methods, and adoption of artificial intelligence (AI) and natural language processing (NLP) are transforming PV by enabling earlier detection and richer characterization of safety signals. Regulatory frameworks (WHO Programme for International Drug Monitoring, FDA FAERS, national programs such as India’s PvPI) continue to evolve to integrate new data sources and analytical tools. Critical challenges remain: data quality and interoperability, algorithm transparency and validation, underreporting, low global PV capacity in many regions, and ethical/privacy concerns. This review synthesizes foundational PV principles, current methods, regulatory context, practical challenges, and future directions including AI, RWD, and global capacity building. 

Keywords: Pharmacovigilance, adverse drug reaction; signal detection; real-world data; artificial intelligence; spontaneous reporting; regulatory surveillance.