ClinicalBench: Can LLMs Beat Traditional ML Models in Clinical Prediction?

TL;DR

This study introduces ClinicalBench, a comprehensive benchmark comparing LLMs and traditional ML models in clinical prediction, revealing that current LLMs do not outperform traditional models in this domain.

Contribution

The paper presents ClinicalBench, a new benchmark for evaluating clinical predictive models, and provides extensive empirical analysis showing LLMs' current limitations compared to traditional ML methods.

Findings

LLMs do not outperform traditional ML models in clinical prediction tasks.

Traditional ML models remain more effective for clinical decision-making.

LLMs show potential but need further development for clinical applications.

Abstract

Large Language Models (LLMs) hold great promise to revolutionize current clinical systems for their superior capacities on medical text processing tasks and medical licensing exams. Meanwhile, traditional ML models such as SVM and XGBoost have still been mainly adopted in clinical prediction tasks. An emerging question is Can LLMs beat traditional ML models in clinical prediction? Thus, we build a new benchmark ClinicalBench to comprehensively study the clinical predictive modeling capacities of both general-purpose and medical LLMs, and compare them with traditional ML models. ClinicalBench embraces three common clinical prediction tasks, two databases, 14 general-purpose LLMs, 8 medical LLMs, and 11 traditional ML models. Through extensive empirical investigation, we discover that both general-purpose and medical LLMs, even with different model scales, diverse prompting or fine-tuning strategies, still cannot beat traditional ML models in clinical prediction yet, shedding light on their potential deficiency in clinical reasoning and decision-making. We call for caution when practitioners adopt LLMs in clinical applications. ClinicalBench can be utilized to bridge the gap between LLMs' development for healthcare and real-world clinical practice.