AI and climate change redefine global disease surveillance

A major new bibliometric study has mapped the global evolution of research on early warning systems for infectious diseases over the past 25 years. Titled “Hotspots and Trends in Research on Early Warning of Infectious Diseases: A Bibliometric Analysis Using CiteSpace” and published in Healthcare, the study systematically analyzes 798 peer-reviewed articles from 1999 to 2024. Using the CiteSpace visualization tool, the research identifies key institutions, emerging themes, and scientific paradigms that now define the field.

The study, conducted by researchers from Nanjing Medical University, highlights an accelerating global shift toward interdisciplinary surveillance systems rooted in artificial intelligence (AI), environmental data, and the One Health approach. While traditional research focused on outbreak monitoring and climate-disease interactions, the current frontier emphasizes real-time predictive tools that integrate human, animal, and ecological health data.

Who is leading global research in infectious disease early warning?

The United States and China are the top contributors to early warning system research, collectively accounting for more than half of all global publications between 1999 and 2024. The United States maintains the highest international collaboration index, while China’s output is largely driven by specialized institutions such as the Chinese Center for Disease Control & Prevention and the Chinese Academy of Sciences.

International organizations such as the World Health Organization (WHO) also hold a central position, particularly in network centrality - a metric used to determine collaborative influence. Notably, most collaboration remains concentrated within developed nations, suggesting the need for more equitable cross-regional partnerships. Top institutions also include Harvard University, the University of California System, and the London School of Hygiene & Tropical Medicine, reflecting the dominance of Western academia in global health surveillance systems.

Author contributions are fragmented, with top producers such as John M. Drake and Wenbiao Hu publishing independently without forming dense collaboration clusters. Meanwhile, the WHO remains the most frequently co-cited entity, emphasizing its global authority in public health guidance and research dissemination.

What are the current hotspots and research frontiers?

The bibliometric keyword analysis reveals several dominant and emerging themes. Earlier research, especially pre-2020, was centered around climate change, influenza, SARS, and surveillance systems. These topics reflect the academic community’s concern with the environmental drivers of disease emergence and the need for robust disease-monitoring infrastructures.

Since the onset of the COVID-19 pandemic, however, the research landscape has shifted dramatically. “SARS-CoV-2”, “wastewater-based epidemiology (WBE)”, “One Health”, and “artificial intelligence” have emerged as the most prominent new keywords, indicating a pivot toward pathogen-specific surveillance and ecosystem-based frameworks. The study notes that WBE gained prominence due to its utility in detecting viral loads, such as SARS-CoV-2, in community sewage systems, serving as an early detection mechanism for population-level infection trends.

Artificial intelligence is increasingly used for real-time outbreak forecasting and early intervention modeling. Machine learning algorithms, deep neural networks, and high-dimensional data integration allow for accurate predictions of infectious disease dynamics, especially in urban settings where traditional monitoring may lag.

Simultaneously, the One Health framework, an integrated surveillance paradigm linking human, animal, and environmental health, has seen rapid adoption in both research and public health policy. This approach is credited with offering a holistic response to zoonotic outbreaks, many of which (e.g., H5N1, H7N9, and SARS-CoV-2) originate from animal reservoirs.

How is the field evolving and what comes next?

According to a keyword burst and timeline analysis, the field has transitioned through three main phases: initial emphasis on climate-disease links (2004–2016), a focus on pathogen-specific surveillance during the COVID-19 era (2016–2022), and a current pivot toward systems-level integration using WBE, AI, and One Health (2022–2024). The evolution reflects a broader public health shift from reactive response to proactive, multi-sectoral monitoring.

The most cited publications in the field further reinforce this trajectory. Two of the top-cited papers pioneered the use of WBE in SARS-CoV-2 detection, validating it as a practical, cost-effective method for pandemic-scale surveillance. Others established AI-based models that forecasted the spread of COVID-19 and assessed intervention strategies with high precision.

The study concludes with strong recommendations: governments and global health bodies should prioritize real-time surveillance infrastructure that integrates WBE and AI tools. Policymakers are urged to institutionalize cross-sector collaboration under the One Health model, and to fund data-sharing platforms that connect human, animal, and environmental health systems.

Importantly, the researchers caution against overlooking ethical considerations. The expansion of AI surveillance must be accompanied by strict governance frameworks to protect privacy and avoid exacerbating health disparities. As AI, climate change, and zoonotic threats converge, the challenge lies not in forecasting the next outbreak, but in being structurally prepared to act on the warnings in real-time.