Skip to content
DraconiticResearch
Search Sources API
186,625 studies31 sourcesUpdated 9 min ago
Peer-reviewedCC BY-NC-NDJournal of translational medicine

Progress in pharmacovigilance research based on artificial intelligence and real-world data: a structured narrative review.

  • Xuelin Sun1,2,3,4
  • Dongfang Qian5
  • Wenjing Zhao5
  • Jingyi Zhang6
  • Pengfei Jin7,8,9,10
  • Yatong Zhang7,8,9,10
10 affiliations
  1. 1Department of Pharmacy, Beijing Hospital, National Center for Gerontology, Beijing, 100730, P.R. China. [email protected].
  2. 2National Clinical Research Center for Gerontology, Beijing, 100730, P.R. China. [email protected].
  3. 3The Key Laboratory of Geriatrics of NHC, Beijing, 100730, P.R. China. [email protected].
  4. 4Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, 100730, P.R. China. [email protected].
  5. 5Beijing North Medical Health Economic Research Center, Beijing, 100020, P.R. China.
  6. 6College of Business Administration, Shenyang Pharmaceutical University, Shenyang, 110016, P.R. China.
  7. 7Department of Pharmacy, Beijing Hospital, National Center for Gerontology, Beijing, 100730, P.R. China.
  8. 8National Clinical Research Center for Gerontology, Beijing, 100730, P.R. China.
  9. 9The Key Laboratory of Geriatrics of NHC, Beijing, 100730, P.R. China.
  10. 10Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, 100730, P.R. China.
Read full textPDFDOICite

Abstract

Background

Traditional pharmacovigilance (PV) relies heavily on spontaneous reporting systems (SRS), which remain vulnerable to under-reporting and limited mechanistic insight. Artificial intelligence (AI) and real-world data (RWD) offer opportunities to strengthen signal detection, risk prediction, and translational safety assessment, but most applications remain at retrospective development or early implementation stages.Methods

We conducted a structured narrative scoping review with critical appraisal, reported using PRISMA-ScR. Literature from 2010 to 2026 was identified through expert review, reference chaining, and regulatory sources. English-language publications on AI or machine learning in clinical pharmacovigilance with real-world data were synthesized thematically across data sources, analytical methods, validation strategies, and implementation challenges.Results

RWD ecosystems-including electronic health records (EHRs), administrative claims, multi-omics, and global SRS-can be integrated with supervised ML, natural language processing (NLP), graph neural networks (GNNs), and federated learning (FL) for diverse PV tasks. Reported area under the receiver operating characteristic curve (AUROC) values often exceed 0.90 in internal validation, yet external validation, calibration, decision-curve analysis, and prospective implementation evidence remain inconsistently reported. Common data models such as the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) and FL architectures improve interoperability and privacy-preserving collaboration. Causal and explainable AI methods are emerging but require explicit assumptions and mechanistic confirmation before regulatory-grade causal claims.Conclusions

AI-enabled PV is transitioning from association-focused surveillance toward prediction, mechanistic interpretation, and implementation science, yet widespread clinical or regulatory adoption requires transparent reporting (e.g., TRIPOD + AI, PROBAST), rigorous external and prospective validation, clear separation of association from causation, clinically interpretable outputs, sustained human oversight, and governance within harmonizing regulatory frameworks.

Related

  1. Bias Mitigation Across the Healthcare Artificial Intelligence Lifecycle: A Structured Narrative Review
    Wenzhi Tseng, Kyla S Holbrook, Radhika L Edpuganti +3
    CureusPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  2. HemOncAgent: an artificial intelligence system for retrieving structured and narrative oncology knowledge
    A. J. Yang, H. A. Zaki, A. Seto +6
    medRxivPreprintCC BYFull textEpidemiology & public healthmedRxiv
  3. Terminologia Anatomica Compliance in Generative Artificial Intelligence Outputs: A Structured Narrative Review and Terminology-Specific Verification Checklist.
    Deepak Sharma, Anaida Singh, Ahmed Mahgoub
    CureusPeer-reviewedCC BYFull textEpidemiology & public healthEurope PMC
  4. Artificial Intelligence-Driven Dietary Assessment Software: A Structured Narrative Review of Technological Advances, Persistent Measurement Error and Barriers to Clinical Translation
    Mohammad Idreesh Khan, Abrar Turki, Shadaid Alanezi +3
    NutrientsPeer-reviewedCC BY 4.0Full textEpidemiology & public healthCrossrefEurope PMC
  5. The application progress of artificial intelligence in the diagnosis of Alzheimer’s disease
    Enqi Liu, Meijia Cheng, Yetao Ju +3
    Frontiers in neurologyPeer-reviewedCC BYFull textPsychiatry & mental healthEurope PMCCrossref

Cite

BibTeXRIS
BibTeX
@article{sun2026progress,
  title = {{Progress in pharmacovigilance research based on artificial intelligence and real-world data: a structured narrative review.}},
  author = {Xuelin Sun and Dongfang Qian and Wenjing Zhao and Jingyi Zhang and Pengfei Jin and Yatong Zhang},
  journal = {Journal of translational medicine},
  year = {2026},
  month = jul,
  doi = {10.1186/s12967-026-08717-7},
  url = {https://europepmc.org/article/MED/42958403},
  note = {CC BY-NC-ND},
}
RIS
TY  - JOUR
TI  - Progress in pharmacovigilance research based on artificial intelligence and real-world data: a structured narrative review.
AU  - Xuelin Sun
AU  - Dongfang Qian
AU  - Wenjing Zhao
AU  - Jingyi Zhang
AU  - Pengfei Jin
AU  - Yatong Zhang
PY  - 2026
DA  - 2026/07/31
JO  - Journal of translational medicine
DO  - 10.1186/s12967-026-08717-7
UR  - https://europepmc.org/article/MED/42958403
AB  - Background  Traditional pharmacovigilance (PV) relies heavily on spontaneous reporting systems (SRS), which remain vulnerable to under-reporting and limited mechanistic insight. Artificial intelligence (AI) and real-world data (RWD) offer opportunities to strengthen signal detection, risk prediction, and translational safety assessment, but most applications remain at retrospective development or early implementation stages.Methods  We conducted a structured narrative scoping review with critical appraisal, reported using PRISMA-ScR. Literature from 2010 to 2026 was identified through expert review, reference chaining, and regulatory sources. English-language publications on AI or machine learning in clinical pharmacovigilance with real-world data were synthesized thematically across data sources, analytical methods, validation strategies, and implementation challenges.Results  RWD ecosystems-including electronic health records (EHRs), administrative claims, multi-omics, and global SRS-can be integrated with supervised ML, natural language processing (NLP), graph neural networks (GNNs), and federated learning (FL) for diverse PV tasks. Reported area under the receiver operating characteristic curve (AUROC) values often exceed 0.90 in internal validation, yet external validation, calibration, decision-curve analysis, and prospective implementation evidence remain inconsistently reported. Common data models such as the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) and FL architectures improve interoperability and privacy-preserving collaboration. Causal and explainable AI methods are emerging but require explicit assumptions and mechanistic confirmation before regulatory-grade causal claims.Conclusions  AI-enabled PV is transitioning from association-focused surveillance toward prediction, mechanistic interpretation, and implementation science, yet widespread clinical or regulatory adoption requires transparent reporting (e.g., TRIPOD + AI, PROBAST), rigorous external and prospective validation, clear separation of association from causation, clinically interpretable outputs, sustained human oversight, and governance within harmonizing regulatory frameworks.
SN  - 1479-5876
ER  -