Large Language Models for Drug Overdose Prediction from Longitudinal Medical Records

2025-04-16

AI Summary

This study investigates the use of large language models (LLMs) to directly predict drug overdose risk from longitudinal patient insurance claims narratives. We propose an end-to-end GPT-4o–based approach supporting both zero-shot prompting and supervised fine-tuning, eliminating the need for manual feature engineering or structured data transformation. Empirically, we demonstrate—for the first time—that an LLM achieves clinically viable risk prediction in a zero-shot setting (AUC = 0.82), significantly outperforming conventional models including XGBoost and random forests. Our key contribution lies in overcoming reliance on labeled datasets and domain-knowledge–driven features, thereby establishing that LLMs possess transferable clinical semantic understanding capabilities. This work introduces a novel paradigm for real-world clinical risk modeling using unstructured medical text.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Computer Vision: Large Vision Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The ability to predict drug overdose risk from a patient's medical records is crucial for timely intervention and prevention. Traditional machine learning models have shown promise in analyzing longitudinal medical records for this task. However, recent advancements in large language models (LLMs) offer an opportunity to enhance prediction performance by leveraging their ability to process long textual data and their inherent prior knowledge across diverse tasks. In this study, we assess the effectiveness of Open AI's GPT-4o LLM in predicting drug overdose events using patients' longitudinal insurance claims records. We evaluate its performance in both fine-tuned and zero-shot settings, comparing them to strong traditional machine learning methods as baselines. Our results show that LLMs not only outperform traditional models in certain settings but can also predict overdose risk in a zero-shot setting without task-specific training. These findings highlight the potential of LLMs in clinical decision support, particularly for drug overdose risk prediction.

Problem

• Predict drug overdose risk from medical records
• Compare LLMs with traditional machine learning methods
• Assess zero-shot and fine-tuned LLM performance

Innovation

• Using GPT-4o for drug overdose prediction
• Leveraging longitudinal insurance claims records
• Evaluating zero-shot and fine-tuned settings