Overview of CHIP 2025 Shared Task 2: Discharge Medication Recommendation for Metabolic Diseases Based on Chinese Electronic Health Records

2025-11-09

AI Summary

Chronic metabolic disease patients in China face high readmission risks due to inadequate medication continuity post-discharge. Method: We formulate a novel multi-label discharge medication recommendation task tailored to Chinese electronic health records (EHRs), integrating large language models (LLMs) with ensemble learning to jointly model heterogeneous clinical text and patient-specific treatment variations. Contribution/Results: We introduce CDrugRed—the first high-quality, publicly available Chinese dataset for metabolic disease medication (5,894 real-world inpatient records)—enabling large-scale model evaluation and LLM-based clinical decision-making research. Evaluated in an open competition with 526 participating teams, our top-performing solution achieved a Jaccard score of 0.5102 and F1-score of 0.6267, demonstrating the effectiveness and generalizability of LLM-driven multi-label medication recommendation in real-world Chinese clinical settings.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Discharge medication recommendation plays a critical role in ensuring treatment continuity, preventing readmission, and improving long-term management for patients with chronic metabolic diseases. This paper present an overview of the CHIP 2025 Shared Task 2 competition, which aimed to develop state-of-the-art approaches for automatically recommending appro-priate discharge medications using real-world Chinese EHR data. For this task, we constructed CDrugRed, a high-quality dataset consisting of 5,894 de-identified hospitalization records from 3,190 patients in China. This task is challenging due to multi-label nature of medication recommendation, het-erogeneous clinical text, and patient-specific variability in treatment plans. A total of 526 teams registered, with 167 and 95 teams submitting valid results to the Phase A and Phase B leaderboards, respectively. The top-performing team achieved the highest overall performance on the final test set, with a Jaccard score of 0.5102, F1 score of 0.6267, demonstrating the potential of advanced large language model (LLM)-based ensemble systems. These re-sults highlight both the promise and remaining challenges of applying LLMs to medication recommendation in Chinese EHRs. The post-evaluation phase remains open at https://tianchi.aliyun.com/competition/entrance/532411/.

Problem

• Automating discharge medication recommendations from Chinese EHR data
• Addressing multi-label challenges and clinical text heterogeneity
• Developing patient-specific treatment plans for metabolic diseases

Innovation

• Using Chinese EHR data for medication recommendation
• Constructing CDrugRed dataset with 5894 hospitalization records
• Applying LLM-based ensemble systems for recommendation