%0 Journal Article
%T Living Systematic Surveillance of Safety Signals from Spontaneous Reports to Federated Health Data
%A Maria Silva
%A Joao Pereira
%A Bruno Costa
%A Ricardo Alves
%J International Journal of Pharmaceutical Research and Allied Sciences
%@ 2277-3657
%D 2026
%V 15
%N 4
%R 10.51847/8fjAi3vVYe
%P 108-119
%X Pharmacovigilance increasingly operates across data environments that differ fundamentally in how exposures, adverse events, populations, denominators, time, and clinical context are represented. Spontaneous reporting remains indispensable for early detection, but signal interpretation increasingly requires longitudinal healthcare data, registries, and distributed analytical networks. Conventional episodic reviews poorly accommodate the continual arrival of new signals, methodological changes, and cross-source evidence. To develop an evidence-bounded living systematic surveillance approach that preserves source-specific inference while enabling recurrent reassessment of drug-safety signals as evidence accumulates across spontaneous reports and longitudinal health data. A living systematic review framework was constructed for peer-reviewed evidence published from 2017 through 2026. The review distinguishes evidence used for signal detection, contextualization, analytical testing, bias assessment, replication, and escalation. Living-review principles govern updating, versioning, eligibility reassessment, and documentation of methodological changes. Evidence is interpreted according to data provenance, exposure definition, adverse-event phenotype, temporal structure, background risk, confounding, clinical seriousness, and cross-source consistency. The evidence indicates that no surveillance source can independently support the full transition from signal emergence to actionable safety inference. Spontaneous reporting is especially sensitive to unexpected patterns but lacks reliable population denominators and is vulnerable to selective reporting. Longitudinal healthcare data offer denominators and temporality but introduce exposure, outcome, indication, and healthcare-recording biases. The synthesis therefore supports source-specific signal states rather than a single hierarchy of evidence. Cross-source agreement strengthens some interpretations but is not itself proof of causality because shared or correlated biases may generate concordance. Living drug-safety surveillance should be organized as a versioned process of detection, testing, reconciliation, and escalation rather than as a one-time search for statistical signals. The proposed architecture preserves distinctions between statistical disproportionality, supported association, causal interpretation, clinical importance, and regulatory actionability.
%U https://ijpras.com/article/living-systematic-surveillance-of-safety-signals-from-spontaneous-reports-to-federated-health-data-no244xpkbwxlsyw