Evolution of Pharmacovigilance Systems: From Spontaneous Reporting to Automated Signal Detection Tools

Pharmacovigilance has developed from paper forms and voluntary reports of suspected adverse drug reactions (ADRs) into a data-driven field integrating electronic case processing, statistical data mining, real-world healthcare data and, progressively, artificial intelligence (AI). Voluntary reporting continues to play a central role because it can reveal unexpected harms across products and patient groups after marketing. However, under-reporting, reporting bias, incomplete details and the absence of exposure denominators restrict its ability to calculate incidence or demonstrate causation. Electronic safety databases made it possible to pool reports at scale and apply consistent coding, while disproportionality methods such as the proportional reporting ratio, reporting odds ratio and Bayesian approaches supported more consistent and reproducible screening. Current systems build on these methods by using duplicate detection, natural language processing, machine learning, temporal analysis and integration of electronic health records, claims, registries and literature. Such tools can rank medicine–event pairs for specialist assessment, reduce repetitive tasks and may shorten the interval between the appearance of a possible risk and its investigation. Even so, an alert generated by an algorithm does not establish a causal relationship. Bias in source data, confounding, stimulated reporting, poor interoperability, limited explainability and inadequate external validation can generate false positives or conceal genuine risks. This review describes key historical developments, major analytical methods, automated workflows, practical uses, advantages, limitations and implementation needs, including the Indian setting and the Pharmacovigilance Programme of India. The review supports a human-led model of automation: computational methods should make screening faster and more consistent, while clinical, epidemiological and regulatory professionals remain accountable for validating signals, evaluating benefit–risk and deciding on action.