Natural Language Processing in Support of Evidence-based Medicine: A Scoping Review

TL;DR

The rapid expansion of medical literature poses a scalability challenge for evidence-based medicine (EBM). This scoping review analyzes 129 studies from 2019 to 2024 to map how natural language processing (NLP) supports the EBM five-step cycle (Ask, Acquire, Appraise, Apply, Assess) through tasks such as information retrieval, entity and relation extraction, quality assessment, synthesis, and summarization, with a focus on transformer-based models and large language models. It catalogs methods, benchmarks, and domain applications, highlighting advances in evidence retrieval, PICO extraction, evidence ranking, meta-analysis assistance, and QA, while identifying critical gaps in datasets for evidence synthesis and question answering and in specialty coverage. The authors also discuss challenges—especially hallucination, attribution, and data requirements—and propose future directions, including few-shot learning, RAG-based systems, and integration of real-world data to enable robust clinical workflows. Overall, the review clarifies how NLP can accelerate evidence synthesis and dissemination, potentially improving the timeliness and reliability of clinical decision-making in diverse settings.

Abstract

Evidence-based medicine (EBM) is at the forefront of modern healthcare, emphasizing the use of the best available scientific evidence to guide clinical decisions. Due to the sheer volume and rapid growth of medical literature and the high cost of curation, there is a critical need to investigate Natural Language Processing (NLP) methods to identify, appraise, synthesize, summarize, and disseminate evidence in EBM. This survey presents an in-depth review of 129 research studies on leveraging NLP for EBM, illustrating its pivotal role in enhancing clinical decision-making processes. The paper systematically explores how NLP supports the five fundamental steps of EBM -- Ask, Acquire, Appraise, Apply, and Assess. The review not only identifies current limitations within the field but also proposes directions for future research, emphasizing the potential for NLP to revolutionize EBM by refining evidence extraction, evidence synthesis, appraisal, summarization, enhancing data comprehensibility, and facilitating a more efficient clinical workflow.

Figures (2)

• Figure 1: PRISMA flow diagram.
• Figure 2: Distribution of papers in different EBM tasks over time. The color schema is the same as Supplementary Table \ref{['tab:studies']}.