arXiv cs.CL· Daisy Rae Lopez, Angela Boccia, Maurizio Sessa·· 12 小时前AI 评分29
MARCA：基于多智能体检索增强的药物警戒个例安全性报告因果关系评估
Multi-Agent Retrieval-Augmented Causality Assessment (MARCA) of Individual Case Safety Reports in Pharmacovigilance
AI 导读
研究团队提出 MARCA 四智能体框架，基于 Gemini 2.5 Pro 与 DeepSeek-V3.2 完成 Naranjo 因果关系评估，在 215 份 ICSR、1,562 个药物-事件组合上与专家共识对照，最终分类一致率 79.4%（kappa = 0.587）。
正文
Abstract:Introduction: Increasing volumes of individual case safety reports (ICSRs) challenge expert-led causality assessment in pharmacovigilance. We developed MARCA (Multi-Agent Retrieval-Augmented Causality Assessment), a four-agent framework for Naranjo-based assessment, and evaluated agreement with human experts and exploratory differences between non-RAG and RAG-labelled scores.
Methods: MARCA combined scientific literature retrieval, drug-label assessment, objective-evidence classification, and clinical causality reasoning using Gemini 2.5 Pro and DeepSeek-V3.2. Agent-generated rationales informed human-assisted scoring. A consensus expert reference standard covered 215 ICSRs and 1,562 drug-event combinations (DECs) from FAERS (Q1-Q2 2025). Agreement was measured using percentage agreement and Cohen's kappa.
Results: Question-level agreement was 96.8% for Naranjo Q2 (kappa = 0.491; 95% CI 0.265-0.663), 85.4% for Q5 (kappa = 0.269; 95% CI 0.180-0.340), and 96.8% for Q10 (kappa = 0.922; 95% CI 0.885-0.951). Final causality classification agreement was 79.4% (95% CI 71.8-85.3; kappa = 0.587, 95% CI 0.450-0.690). In a retrospective subset of 60 DECs, agreement was higher for RAG-labelled than non-RAG scores for Q5 (83.3% vs 65.0%) and Q10 (100.0% vs 18.3%).
Discussion: MARCA-assisted assessment showed substantial variation in agreement across Naranjo questions, with lower concordance for alternative causes. Retrospective RAG comparisons are exploratory and do not establish prospective effectiveness.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.33279 [cs.CL] |
| (or arXiv:2610.33279v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33279 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Maurizio Sessa Dr. [view email]
[v1]
Fri, 23 Oct 2026 20:43:29 UTC (2,619 KB)
来源：arXiv cs.CL · arxiv.org