Large language models in global health

Large language models (LLMs) are emerging as powerful tools in healthcare, with a growing role in global health, particularly in low- and middle-income countries (LMICs). This Perspective examines the current progress, challenges and prospects of LLMs in addressing health system disparities and supporting the achievement of the Sustainable Development Goals (SDGs). While high-income countries dominate the development and deployment of LLMs, LMICs face substantial barriers. These include limited digital infrastructure, a scarcity of locally relevant data, regulatory gaps, under-representation of local languages and dialects, and challenges related to privacy and data security. The limited availability of local expertise, capacity building programmes and sustained technical support remains a key barrier to scaling LLMs in LMICs. Nonetheless, case studies highlight how mobile-based LLM applications, hybrid artificial intelligence systems and open-weight models like DeepSeek are enhancing access to care, improving diagnostics and supporting clinical decision-making in resource-limited settings. Key risks include model hallucinations, equity concerns and environmental impacts. These underscore the need for rigorous validation, localized fine-tuning and global governance frameworks. The implementation of LLMs with contextual sensitivity, responsible oversight and codevelopment partnerships is important to avoid perpetuating health inequities. With the right safeguards and strategic investments in capacity building, LLMs have the potential to transform global health by bridging divides in access, augmenting overburdened health workforces, and enabling scalable and cost-effective innovations for the most underserved communities. Large language models (LLMs) are emerging as powerful tools in healthcare, with a growing role in global health, particularly in low- and middle-income countries. This Perspective examines the current progress, challenges and prospects of LLMs in addressing health system disparities and supporting achievement of the Sustainable Development Goals.

Authors

Institutions

Publication Details

Journal
Nature Health
Published
2026-01-15
DOI
https://doi.org/10.1038/s44360-025-00024-7
Citations
4
Primary Topic
Global Health and Surgery
Type
article
Field-Weighted Citation Impact
167.39

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Large language models in global health

Kwesi Nyan Amissah-Arthur, Siegfried K. Wagner, 弘道 新井川, Gary Collins et al.
4 citations
Nature Health
Global Health and Surgery
167.39
article

Large language models in global health

Kwesi Nyan Amissah-Arthur, Siegfried K. Wagner, 弘道 新井川, Gary Collins, Wendy W. Chapman, Benjamin Alan Goldstein, Bilal Akhter Mateen, Eric J. Topol, Xiaoxuan Liu, Iain Bee Huat Tan, Michelle María Jiménez de Tavárez, Yih Chung Tham, Bin Sheng, Josip Car, Marcus Eng Hock Ong, Nan Liu, Danielle S. Bitterman, Jasmine Chiat Ling Ong, Tien Yin Wong, Pearse A. Keane, Yun Liu, Rui Yang, Alastair K. Denniston, Hiang Khoon Tan, Karel G. M. Moons, Yilin Ning, Phuoc V. Le, Lionel Tim-Ee Cheng
article en
4 citations

Abstract

Large language models (LLMs) are emerging as powerful tools in healthcare, with a growing role in global health, particularly in low- and middle-income countries (LMICs). This Perspective examines the current progress, challenges and prospects of LLMs in addressing health system disparities and supporting the achievement of the Sustainable Development Goals (SDGs). While high-income countries dominate the development and deployment of LLMs, LMICs face substantial barriers. These include limited digital infrastructure, a scarcity of locally relevant data, regulatory gaps, under-representation of local languages and dialects, and challenges related to privacy and data security. The limited availability of local expertise, capacity building programmes and sustained technical support remains a key barrier to scaling LLMs in LMICs. Nonetheless, case studies highlight how mobile-based LLM applications, hybrid artificial intelligence systems and open-weight models like DeepSeek are enhancing access to care, improving diagnostics and supporting clinical decision-making in resource-limited settings. Key risks include model hallucinations, equity concerns and environmental impacts. These underscore the need for rigorous validation, localized fine-tuning and global governance frameworks. The implementation of LLMs with contextual sensitivity, responsible oversight and codevelopment partnerships is important to avoid perpetuating health inequities. With the right safeguards and strategic investments in capacity building, LLMs have the potential to transform global health by bridging divides in access, augmenting overburdened health workforces, and enabling scalable and cost-effective innovations for the most underserved communities. Large language models (LLMs) are emerging as powerful tools in healthcare, with a growing role in global health, particularly in low- and middle-income countries. This Perspective examines the current progress, challenges and prospects of LLMs in addressing health system disparities and supporting achievement of the Sustainable Development Goals.

Nature HealthVol. 1(1)
Brigham and Women's Hospital (US), Google (United States) (US), Program for Appropriate Technology in Health (US), University Hospitals Birmingham NHS Foundation Trust (GB), Moorfields Eye Hospital NHS Foundation Trust (GB), Harvard University (US), University of Ghana (GH), Scripps Institution of Oceanography (US), SingHealth (SG), Pontificia Universidad Católica Madre y Maestra (DO), The University of Melbourne (AU), National University of Singapore (SG), Duke University (US), Shanghai Jiao Tong University (CN), King's College London (GB), Utrecht University (NL), Singapore General Hospital (SG), Singapore National Eye Center (SG), Korle Bu Teaching Hospital (GH), University Medical Center Utrecht (NL), University College Birmingham (GB), Singapore Eye Research Institute (SG), Dana-Farber Cancer Institute (US), Duke Institute for Health Innovation (US), Duke-NUS Medical School (SG), National Cancer Centre Singapore (SG), Scripps (United States) (US), VinUniversity (VN), National University Health System (SG), Moorfields Eye Hospital (GB), Beijing Tsinghua Chang Gung Hospital (CN), Dana-Farber Brigham Cancer Center (US), NIHR Birmingham Biomedical Research Centre (GB), University College London (GB), Mass General Brigham (US), University of Birmingham (GB), Tsinghua University (CN)
Patient-Centered Outcomes Research Institute, Duke-NUS Medical School, National Natural Science Foundation of China
Partnerships for the goals
Openalex Percentile: Top 0%
Global Health and Surgery
167.39
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.