Medical research production in native languages: A descriptive analysis of PubMed database

Abstract

INTRODUCTION: Language barriers in medicine can hinder effective communication, comprehension, and patient care. While English has emerged as the dominant language in global medicine, the importance of native languages should not be overlooked. This article aims to examine the extent of publishing in native languages by analyzing the PubMed database literature to gain further insights into the usage of native languages in medicine and medical research. METHODS: In December 2023, a comprehensive examination of the PubMed literature was conducted for each of the 55 registered languages. We searched for records published in each language (e.g., German[lang]) by applying language filters. Ethnologue provided data on the number of worldwide native speakers for each language, facilitating a comparative analysis. RESULTS: By December 2023, PubMed contained over 36 million publications, with 86.5% of them published in English. German, French, and Russian came after English, with over 700 thousand publications each. Among the languages analyzed, fourteen had fewer than 50 publications, nineteen had fewer than 100, twenty-two had fewer than 500, and twenty-five had fewer than one thousand publications. European languages were well-represented with thousands of publications each, while widely spoken languages such as Hindi and Arabic had limited representation. CONCLUSION: The production of medical research in native languages reflects the attention given to native languages in medicine and medical education within each country. It is crucial to provide due attention to these language-related issues and explore strategies for including native languages in medicine to bridge the gaps in language and medicine.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 10 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.