Characterising global antimicrobial resistance research explains why One Health solutions are slow in development: An application of AI-based gap analysis
Cai Chen a b 1, Shu-Le Li a b 1, Yao-Yang Xu a c, Jue Liu d e, David W. Graham f, Yong-Guan Zhu a c g
https://doi.org/10.1016/j.envint.2024.108680
Highlights
- A global dataset with 254,738 AMR-related manuscripts was complied.
- An AI information retrieval system based on NLP was developed to extract knowledge.
- Global foci and current knowledge gaps of AMR research were characterized.
- More intersectoral AMR research is urgently needed to guide One Health AMR solutions.
- Intrasectoral and intersectoral AMR research co-occurrence networks were constructed.
Abstract
The global health crisis posed by increasing antimicrobial resistance (AMR) implicitly requires solutions based a One Health approach, yet multisectoral, multidisciplinary research on AMR is rare and huge knowledge gaps exist to guide integrated action. This is partly because a comprehensive survey of past research activity has never performed due to the massive scale and diversity of published information. Here we compiled 254,738 articles on AMR using Artificial Intelligence (AI; i.e., Natural Language Processing, NLP) methods to create a database and information retrieval system for knowledge extraction on research perfomed over the last 20 years. Global maps were created that describe regional, methodological, and sectoral AMR research activities that confirm limited intersectoral research has been performed, which is key to guiding science-informed policy solutions to AMR, especially in low-income countries (LICs). Further, we show greater harmonisation in research methods across sectors and regions is urgently needed. For example, differences in analytical methods used among sectors in AMR research, such as employing culture-based versus genomic methods, results in poor communication between sectors and partially explains why One Health-based solutions are not ensuing. Therefore, our analysis suggest that performing culture-based and genomic AMR analysis in tandem in all sectors is crucial for data integration and holistic One Health solutions. Finally, increased investment in capacity development in LICs should be prioritised as they are places where the AMR burden is often greatest. Our open-access database and AI methodology can be used to further develop, disseminate, and create new tools and practices for AMR knowledge and information sharing.
CHAT GPT 4 READS CHEN 2024
Overview
This study aims to understand why solutions to antimicrobial resistance (AMR) using the One Health approach have been slow to develop. It uses artificial intelligence (AI) to analyze a large amount of research data on AMR to identify knowledge gaps and propose ways to enhance research and policy-making.
Key Points
The Problem of AMR:
- AMR is a major global health issue, leading to significant morbidity and mortality.
- The spread of AMR is exacerbated by the interconnectedness of humans, animals, and the environment.
One Health Approach:
- One Health is an interdisciplinary strategy that recognizes the interconnectedness of human, animal, and environmental health.
- Effective AMR solutions require integrated actions across these sectors.
Research Gaps:
- Despite the importance of a One Health approach, there is limited multisectoral and multidisciplinary research on AMR.
- There is a lack of comprehensive surveys of past research due to the vast amount of published information.
AI-based Analysis:
- The study used AI and natural language processing (NLP) to analyze 254,738 articles on AMR published over the last 20 years.
- AI helped create a database and an information retrieval system to extract knowledge from these articles.
- The analysis identified significant gaps in intersectoral research, especially in low-income countries (LICs).
Findings:
- There is a need for harmonization of research methods across different sectors.
- Differences in analytical methods (e.g., culture-based vs. genomic methods) hinder effective communication and data integration.
- Increased investment in capacity development in LICs is crucial as they bear a significant burden of AMR.
Research Patterns:
- AMR research is concentrated in high-income countries (HICs), with fewer studies in LICs.
- The pathogens most commonly studied include E. coli, S. aureus, and Salmonella spp.
- Methods like antimicrobial susceptibility testing (AST) and whole genomic sequencing (WGS) are increasingly used, but there are disparities in method usage across regions.
Policy Implications:
- The study suggests that better integration of different research methods and more focus on cross-sectoral research are needed.
- Emphasizes the importance of coordinated global efforts to address AMR.
Conclusion
The study highlights the importance of using AI to synthesize and analyze large amounts of research data to identify gaps and guide future research. It stresses the need for more integrated and harmonized research efforts to effectively tackle the AMR crisis through a One Health approach.
This summary should provide you with a clearer understanding of the main ideas and findings of the article.
PLEASE, DEEPER DIVE INTO FINDINGS
Certainly! Let's dive deeper into the findings of the article "Characterising global antimicrobial resistance research explains why One Health solutions are slow in development: An application of AI-based gap analysis" by Cai Chen et al.
Detailed Findings
Research Concentration and Regional Disparities:
- High-Income Countries (HICs):
- Most AMR research is concentrated in HICs. The majority of publications come from countries like the United States, China, and European nations.
- HICs have better resources and infrastructure for conducting comprehensive AMR research, including advanced methodologies like whole genomic sequencing (WGS) and metagenomics.
- Low-Income Countries (LICs):
- LICs have significantly fewer AMR publications. The studies that do exist often rely on more basic methodologies such as PCR-based methods.
- There is a notable lack of capacity and infrastructure in LICs to conduct advanced AMR research, which hampers the ability to develop comprehensive solutions.
- LICs suffer from a high burden of AMR, but the research output does not match the need due to limited resources.
- High-Income Countries (HICs):
Pathogens and Regions:
- The study found that certain pathogens are more frequently studied in specific regions:
- E. coli, S. aureus, Pseudomonas aeruginosa, Salmonella spp., and Klebsiella pneumoniae are the top five reported pathogens globally.
- In South Asia, Salmonella typhi is particularly prevalent due to the high incidence of typhoid fever.
- East Asia and the Pacific see significant research on Salmonella spp., Acinetobacter spp., Shigella spp., and Neisseria gonorrhoeae.
- The study found that certain pathogens are more frequently studied in specific regions:
Pathogen-Drug Combinations:
- Research on specific pathogen-drug combinations varies by region and income level:
- Methicillin-resistant Staphylococcus aureus (MRSA) is the most studied pathogen-drug combination globally.
- ESBL-producing E. coli and carbapenem-resistant Enterobacteriaceae are also highly studied.
- In regions like Europe and Central Asia, research on carbapenem-resistant and ESBL-producing Enterobacteriaceae is more common.
- Latin America and the Caribbean have the lowest number of related publications.
- Research on specific pathogen-drug combinations varies by region and income level:
Antibiotic Resistance Genes (ARGs):
- The distribution of research on ARGs varies significantly:
- mecA is the most frequently studied ARG, with the highest number of publications coming from Iran, Brazil, the United States, Egypt, and China.
- Different countries focus on different ARGs based on local relevance and research priorities. For example, China focuses on gyrA, sul1, sul2, and parC, while India has significant research on blaNDM-1.
- The co-occurrence of ARGs and pathogens in publications is common, with E. coli frequently associated with ARGs like ampC and mcr-1.
- The distribution of research on ARGs varies significantly:
Research Methods and Trends:
- The study highlights a dramatic increase in the use of advanced methods like WGS and metagenomics over the last decade:
- AST (Antimicrobial Susceptibility Testing) and PCR-based methods are the most commonly used methodologies across all income levels.
- WGS usage has surged, especially in HICs, but remains underutilized in LICs and LMICs.
- Metagenomics has also become more prevalent, particularly in environmental research.
- There is a need for harmonization of methods across sectors to improve data integration and communication.
- The study highlights a dramatic increase in the use of advanced methods like WGS and metagenomics over the last decade:
One Health Sector Analysis:
- Research is heavily concentrated in the human health sector, but there is increasing interest in other sectors:
- Environmental and animal health sectors have seen significant growth in AMR research.
- Research on the interface of human health and environment is the most common inter-sectoral activity, followed by the interface of animal health and food systems.
- Plant health research, while growing, remains less common compared to other sectors.
- Research is heavily concentrated in the human health sector, but there is increasing interest in other sectors:
Stages of AMR Research:
- The study identifies three stages of AMR research development from 2003 to 2022:
- Stage 1 (2003-2011): Initial growth in AMR research with a focus on traditional methods.
- Stage 2 (2012-2018): Expansion of research with more inter-sectoral studies and the introduction of advanced methods like WGS.
- Stage 3 (2019-2022): Rapid growth in the use of advanced methods and a significant increase in multi-sectoral research.
- The study identifies three stages of AMR research development from 2003 to 2022:
Investment and Capacity Building:
- There is a critical need for increased investment in AMR research, especially in LICs:
- Enhancing research capacity and infrastructure in LICs is essential to address the high burden of AMR.
- Greater investment in advanced methodologies and harmonization of research practices across regions and sectors will facilitate better data integration and policy-making.
- There is a critical need for increased investment in AMR research, especially in LICs:
Conclusion
The study underscores the importance of using AI to analyze vast amounts of research data to identify gaps and guide future research efforts. It highlights the disparities in AMR research across different regions and income levels and emphasizes the need for integrated, harmonized, and well-resourced research efforts to effectively address the global AMR crisis through a One Health approach.
