Thousands of dengue forecasting models have been published worldwide, but very few have been tested in real public health settings. Now, researchers from Australia, the United States and Vietnam are moving beyond theory, launching a major field evaluation to determine whether a new early-warning platform can help health authorities act earlier against a disease that the World Health Organisation says puts nearly half the world's population at risk.
/prod01/channel_8/media/scu-dep/news/images/2025/E-dengue-low-res.jpg)
Play video Video transcript
Every year, dengue outbreaks stretch health systems across tropical regions to their limits. When hospitals begin to fill and communities fall ill, the virus has often already spread too far. For local health teams, that means reacting late – working harder, with fewer results.
But what if we could see dengue risk coming months in advance, and act before the first case appears? What if communities could prepare, instead of being caught off guard?
The Dengue Early Warning Tool makes that vision a reality. Powered by advanced statistical models and artificial intelligence, it integrates years of dengue case reports, detailed climate data, and patterns linked to everyday life habits—such as water storage practices, hygiene behaviours, and local business activity—to forecast where outbreaks are likely to occur, up to three months in advance.
With this insight, national and local teams gain the time they need to act early, plan ahead, and protect communities before dengue takes hold.
What makes this system truly different is how well it fits into real life. It doesn't replace existing systems – it strengthens them. The tool has been designed with users and for users, ensuring it is not only easy to use but genuinely useful in practice—intuitive, aligned with existing workflows, and built to support real decision-making so that it becomes a tool that is truly used by frontline staff.
The tool provides direct support to the national dengue surveillance system and is fully aligned with local reporting schedules and local public health decision-making processes.
Whether in a busy city or a rural health post, online or offline, it's designed for the way people actually work. Built with local health workers, it adapts to every district – supporting collaboration, not confusion, and ensuring no one is left behind.
When risk levels rise, timely alerts go straight to local teams, guiding targeted mosquito control, community engagement, and efficient resource use. In pilot provinces, these early actions have reduced outbreaks, saved time and money, and strengthened trust between communities and health authorities. The tool turns complex data into clear decisions – and decisions into meaningful action.
The Dengue Early Warning Tool is more than technology – it's a bridge between science, policy, and community resilience. As climate change reshapes disease patterns, predictive tools like this are essential for protecting lives and strengthening public health systems.
Together, we can move from reacting to predicting, from crisis to prevention – building a future where dengue no longer catches us by surprise.
Dengue Early Warning Tool – Predict. Prepare. Protect.
For years, scientists have been developing increasingly sophisticated models to forecast dengue outbreaks, drawing on climate data, disease surveillance records and population trends. Despite this growing body of research, most models remain confined to academic journals, never tested in the environments where outbreaks actually unfold.
That gap between prediction and practice is now the focus of a major international study evaluating E-Dengue , a digital early-warning platform designed to support dengue prevention and response at the district level. The system is being introduced in Vietnam as part of one of the largest real-world evaluations of an early warning system for dengue fever to date.
Southern Cross University is leading the second phase of this multi-year collaboration, working alongside the University of Queensland, Yale University and Vietnam's National Institute of Hygiene and Epidemiology. This stage has focused on translating predictive modelling into E-Dengue – an open-source, user-friendly software system designed specifically for frontline public-health use.
Like a highly sophisticated weather forecast, but for dengue
Southern Cross University researcher Dr Vinh Bui said the priority has been ensuring the platform works under real-world conditions.
"There are thousands of published studies on dengue prediction models, but very few become tools that are practical for local teams," said Dr Bui.
"Think of E-Dengue as a highly sophisticated weather forecast, but for dengue. It takes huge amounts of information – like rainfall, temperature and historical case data – and processes it with advanced statistical models and AI.
"Our goal in this stage has been to build a tool that is reliable, actionable, fast and intuitive – something that supports, rather than complicates, routine public health work."
E-Dengue integrates multiple data streams – including historical dengue case data, climate variables such as rainfall and temperature, and local socio-ecological information – to forecast dengue risk one to three months in advance. Rather than producing complex outputs, the platform presents predictions in a clear, operational format that can be used by district-level health authorities to plan surveillance, allocate resources and prepare targeted interventions.
With the predictive models developed and the E-Dengue platform built, the project is now entering its most critical stage: embedding the tool into Vietnam's routine dengue surveillance systems and launching a large cluster randomised controlled trial to test whether earlier warnings lead to earlier action and fewer outbreaks.
"We've built a tool with strong potential, but the critical test is ahead of us," said Dr Bui.
"The next three years will tell us whether early warnings lead to earlier, better-targeted interventions – and whether this improves health outcomes."
This work is guided by the research team's recently published "Useful, Usable, Used (3U) Framework" in Nature Communications, which examines how digital prediction tools can move from innovation into sustained real-world use.
Yale University researcher Dr Robert Dubrow said the next phase of the collaboration will provide rare evidence on whether early-warning systems can shift dengue control from a reactive to a proactive approach.
"Our team at Yale has led the development of the predictive model underpinning the platform," Dr Dubrow said.
"We now look forward to working with our Vietnamese and Australian partners to rigorously evaluate whether early warnings change outcomes in practice."
Interest in the approach is already emerging from neighbouring countries, including Thailand, Laos and Cambodia, where dengue risk is increasing under climate and population pressures.
Full deployment of E-Dengue across selected districts in Vietnam's Mekong Delta region will begin in early 2026. From 2026 to 2028, the platform will be used in day-to-day public-health decision-making while the research team conducts the randomised controlled trial and associated studies.
"This is a challenging and complex process," said University of Queensland Associate Professor Dung Phung.
"Our long-term aim is to develop a tool that Vietnam's Ministry of Health sees value in maintaining beyond the life of the project."
