Artificial Intelligence in Pharmacovigilance: Evaluating Opportunities for Automation, Human Oversight Requirements, and Limitations of Autonomous Decision-Making

Autores/as

  • Harshavardhan Gangavarapu United Therapeutics Author

Palabras clave:

pharmacovigilance, artificial intelligence, human oversight, drug safety, signal detection, automation governance, risk-stratified validation

Resumen

Background: Artificial intelligence (AI) may accelerate pharmacovigilance, but the point at which efficiency gains become incompatible with accountable safety decisions remains uncertain.

Objective: To compare automation, oversight intensity, decision quality, and residual risk across pharmacovigilance tasks in a matched scenario-based benchmark.

Methods: A structured dataset contained 4,096 evaluation records comprising 1,024 scenarios repeated under autonomous AI, human-on-the-loop, human-in-the-loop, and human-in-command arrangements. Sixteen tasks were grouped into five workflow stages. Final correctness, residual high-impact errors, predefined autonomy eligibility, policy adherence, time, and modeled cost were assessed at the scenario level.

Results: Final decision correctness was 77.05% (789/1,024) with autonomous AI, 87.50% (896/1,024) with human-on-the-loop oversight, 94.53% (968/1,024) with human-in-the-loop verification, and 96.00% (983/1,024) under human-in-command governance. Residual high-impact errors affected 12.30%, 6.15%, 3.12%, and 1.46% of scenarios, respectively. Aggregate time savings versus an assigned manual baseline ranged from 95.15% for autonomy to 17.68% for human-in-command. Only 13 scenarios (1.27%) were coded as eligible for autonomous use.

Conclusion: The benchmark reveals a pronounced efficiency–safety trade-off and supports risk-stratified automation of bounded processing tasks, with stronger human control for interpretive and consequential decisions. These scenario-based findings are not evidence of real-world effectiveness or regulatory compliance without external validation.

Referencias

L. Hazell and S. A. W. Shakir, “Under-reporting of adverse drug reactions: a systematic review,” Drug Saf, vol. 29, no. 5, pp. 385–396, 2006, doi:10.2165/00002018-200629050-00003.

O. Caster, K. Juhlin, S. Watson, and G. N. Norén, “Improved statistical signal detection in pharmacovigilance by combining multiple strength-of-evidence aspects in vigiRank,” Drug Saf, vol. 37, no. 8, pp. 617–628, 2014, doi:10.1007/s40264-014-0204-5.

J. K. Aronson, “Artificial intelligence in pharmacovigilance: an introduction to terms, concepts, applications, and limitations,” Drug Saf, vol. 45, no. 5, pp. 407–418, 2022, doi:10.1007/s40264-022-01156-5.

G. L. Martin, J. Jouganous, R. Savidan, et al., “Validation of artificial intelligence to support the automatic coding of patient adverse drug reaction reports, using nationwide pharmacovigilance data,” Drug Saf, vol. 45, no. 5, pp. 535–548, 2022, doi:10.1007/s40264-022-01153-8.

T. Botsis and K. Kreimeyer, “Improving drug safety with adverse event detection using natural language processing,” Expert Opin Drug Saf, vol. 22, no. 8, pp. 659–668, 2023, doi:10.1080/14740338.2023.2228197.

A. L. Silverman, M. Sushil, B. Bhasuran, et al., “Algorithmic identification of treatment-emergent adverse events from clinical notes using large language models: a pilot study in inflammatory bowel disease,” Clin Pharmacol Ther, vol. 115, no. 6, pp. 1391–1399, 2024, doi:10.1002/cpt.3226.

S. Modi, K. A. Kasmiran, N. M. Sharef, and M. Y. Sharum, “Extracting adverse drug events from clinical notes: a systematic review of approaches used,” J Biomed Inform, vol. 151, Art. no. 104603, 2024, doi:10.1016/j.jbi.2024.104603.

M. K. Desai, “Artificial intelligence in pharmacovigilance: opportunities and challenges,” Perspect Clin Res, vol. 15, no. 3, pp. 116–121, 2024, doi:10.4103/picr.picr_290_23.

E. Kopacheva, A. Lincke, O. Björneld, and T. Hammar, “Detecting adverse drug events in clinical notes using large language models,” Stud Health Technol Inform, vol. 327, pp. 892–893, 2025, doi:10.3233/SHTI250495.

P. M. Cutroneo, D. Sartori, M. Tuccori, et al., “Conducting and interpreting disproportionality analyses derived from spontaneous reporting systems,” Front Drug Saf Regul, vol. 3, Art. no. 1323057, 2023, doi:10.3389/fdsfr.2023.1323057.

M. Fusaroli, D. Sartori, E. P. van Puijenbroek, and G. N. Norén, “Charting and sidestepping the pitfalls of disproportionality analysis,” Drug Saf, vol. 49, no. 2, pp. 147–162, 2026, doi:10.1007/s40264-025-01604-y.

E. Scosyrev, S. Behr, D. Jain, A. Ponnuru, and C. Michel, “Disproportionality analysis and causal inference in drug safety,” Pharmaceut Med, vol. 39, no. 2, pp. 97–107, 2025, doi:10.1007/s40290-024-00549-4.

N. Chalabianloo, F. Ahmadi, M. A. Omrani, et al., “Machine learning methods for predicting adverse drug events: a systematic review,” Br J Clin Pharmacol, vol. 92, no. 2, pp. 422–444, 2026, doi:10.1002/bcp.70377.

A. Nagar, J. Gobburu, and A. Chakravarty, “Artificial intelligence in pharmacovigilance: advancing drug safety monitoring and regulatory integration,” Ther Adv Drug Saf, vol. 16, Art. no. 20420986251361435, 2025, doi:10.1177/20420986251361435.

J. Warner, L. H. De Souza Teodoro, A. P. Jardim, and C. Albera, “Intelligent automation improved efficiency in pharmacovigilance safety signal assessment,” Clin Pharmacol Ther, 2026, doi:10.1002/cpt.70409.

D. Ludwig, M. Wang, J. Buchanan, and T. Trinh, “A large language model for extracting post-marketing adverse drug events from clinical notes in the electronic health record,” Drug Saf, vol. 49, no. 10, pp. 1181–1198, 2026, doi:10.1007/s40264-026-01682-6.

A. Franchini, R. Noseda, J. Cornelius, et al., “NLP in support of pharmacovigilance: QUality Adverse Drug Reaction AcTIve Control (QUADRATIC),” Clin Pharmacol Ther, vol. 119, no. 6, pp. 1597–1603, 2026, doi:10.1002/cpt.70250.

J. M. Plasek, Y. Li, M. G. Amato, et al., “Relationship extraction for adverse drug events in clinical notes using large language models: a retrospective cross-sectional study,” EBioMedicine, vol. 133, Art. no. 106516, 2026, doi:10.1016/j.ebiom.2026.106516.

S. G. Lim and M. A. Ismail, “Automated multi-label adverse drug reaction extraction from Malaysia’s national pharmacovigilance narratives,” J Med Syst, vol. 50, Art. no. 120, 2026, doi:10.1007/s10916-026-02448-4.

A. Katz, A. Dhankar, G. Fransoo, et al., “Using natural language processing to identify adverse drug events characterized by medication replacement in primary care electronic medical records: algorithm and validation study,” J Med Internet Res, vol. 28, Art. no. e86467, 2026, doi:10.2196/86467.

I. El Sayed, R. N. Bedwany, M. Salama, and A. S. H. Eltrass, “Integration of artificial intelligence into pharmacovigilance: a systematic review,” Res Social Adm Pharm, 2026, doi:10.1016/j.sapharm.2026.09.001.

T. Y. Kim, W. S. Oh, and D. H. Jeong, “Large language models in adverse drug reaction detection and pharmacovigilance: a systematic review of current applications, challenges, and future directions,” Diagnostics, vol. 16, no. 15, Art. no. 2435, 2026, doi:10.3390/diagnostics16152435.

G. N. Norén, J. Durand, V. Kara, et al., “Artificial intelligence in pharmacovigilance: guiding principles from the CIOMS Working Group XIV,” Drug Saf, 2026, doi:10.1007/s40264-026-01708-z.

L. Wu, J. Xu, O. Dang, and R. Ball, “Does generative AI mean the “end of history” for pharmacovigilance automation? Towards a framework for the future of human–AI systems,” Front Drug Saf Regul, vol. 6, Art. no. 1846339, 2026, doi:10.3389/fdsfr.2026.1846339.

G. S. Collins, K. G. M. Moons, P. Dhiman, et al., “TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods,” BMJ, vol. 385, Art. no. e078378, 2024, doi:10.1136/bmj-2023-078378.

X. Liu, S. Cruz Rivera, D. Moher, M. J. Calvert, and A. K. Denniston, “Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension,” Nat Med, vol. 26, no. 9, pp. 1364–1374, 2020, doi:10.1038/s41591-020-1034-x.

S. Cruz Rivera, X. Liu, A. W. Chan, et al., “Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension,” Nat Med, vol. 26, no. 9, pp. 1351–1363, 2020, doi:10.1038/s41591-020-1037-7.

C. J. Kelly, A. Karthikesalingam, M. Suleyman, G. Corrado, and D. King, “Key challenges for delivering clinical impact with artificial intelligence,” BMC Med, vol. 17, Art. no. 195, 2019, doi:10.1186/s12916-019-1426-2.

K. Lekadir, A. F. Frangi, A. R. Porras, et al., “FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare,” BMJ, vol. 388, Art. no. e081554, 2025, doi:10.1136/bmj-2024-081554.

M. Hauben, “Artificial intelligence in pharmacovigilance: do we need explainability? Pharmacoepidemiol Drug Saf,” 2022;31(12):1311–1316, doi:10.1002/pds.5501.

M. Ghassemi, L. Oakden-Rayner, and A. L. Beam, “The false hope of current approaches to explainable artificial intelligence in health care,” Lancet Digit Health, vol. 3, no. 11, pp. e745–e750, 2021, doi:10.1016/S2589-7500(21)00208-9.

Z. Obermeyer, B. Powers, C. Vogeli, and S. Mullainathan, “Dissecting racial bias in an algorithm used to manage the health of populations,” Science, vol. 366, no. 6464, pp. 447–453, 2019, doi:10.1126/science.aax2342.

European Medicines Agency, “Reflection paper on the use of artificial intelligence (AI) in the medicinal product lifecycle,” EMA/CHMP/CVMP/83833/2023. 2024. https://www.ema.europa.eu/en/scientific-guidelines/use-artificial-intelligence-ai-medicinal-product-lifecycle.

European Parliament and Council, “Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act),” Off J Eur Union. 2024. https://eur-lex.europa.eu/eli/reg/2024/1689/oj.

World Health Organization, “Ethics and governance of artificial intelligence for health: WHO guidance,” Geneva: WHO; 2021. ISBN 9789240029200. https://www.who.int/publications/i/item/9789240029200.

Descargas

Publicado

2026-11-21

Cómo citar

Artificial Intelligence in Pharmacovigilance: Evaluating Opportunities for Automation, Human Oversight Requirements, and Limitations of Autonomous Decision-Making. (2026). Journal of Evidence-Based Medical Research, 2(03), 105-114. https://jebmr.com/index.php/home/article/view/50

Artículos similares

1-10 de 25

También puede Iniciar una búsqueda de similitud avanzada para este artículo.