BSC leads development of an early warning system that predicts dengue outbreaks in the Caribbean months in advance

In an advance for climate-informed disease early warning systems, a team of researchers, public health actors and meteorologists have developed a prediction model capable of forecasting dengue outbreaks in the Caribbean up to three months in advance. The study, led by Chloe Fletcher and Rachel Lowe, from the Global Health Resilience group at the Barcelona Supercomputing Center-Centro Nacional de Supercomputación (BSC-CNS), introduces a predictive model that captures the complex interplay between drought, heat, and extreme rainfall conditions on dengue outbreak risk in Barbados.

These extreme weather events are becoming more frequent and intense in the Caribbean due to climate change. At the same time, dengue outbreaks and other mosquito-transmitted diseases, such as Chikungunya and Zika, have become increasingly common and explosive across the region over the last 15 years.

Fuelled by climate change, the global incidence of dengue has also increased significantly over the past two decades, posing a major public health challenge. In 2024, 14.1 million dengue cases were reported worldwide, surpassing the historic milestone of 7 million observed in 2023, where over 92% of cases were recorded in the Americas. Significant outbreaks also occurred in Southeast Asia and Africa and even Europe has experienced sporadic local transmission in countries like Italy, France, and Spain, indicating a growing risk of establishment as mosquito vectors expand across the continent.

The effect of successive extreme weather events on dengue outbreak risk

The model presented in the study, published in the prestigious journal The Lancet Planetary Health, incorporates interactions between climate variables, such as temperature and precipitation indicators at varying time lags, to predict the probability of a dengue outbreak and the approximate number of cases in Barbados three months in advance. Specifically, the research found that extremely dry conditions 5 months prior to an outbreak, followed by warmer temperatures 3 months prior, and heavy rainfall 1 month prior, led to the highest incidence of dengue. Using this approach, the model correctly identified 81% of observed outbreaks during model validation from 2012 to 2022, significantly outperforming traditional surveillance models.

"This modelling approach allows us to account for the effect of successive extreme weather events on disease outbreak risk," said Chloe Fletcher, first author of the study and PhD candidate at the BSC’s Global Health Resilience group. “By focusing on interactions between climatic drivers, we were able to better anticipate dengue outbreak risk in Barbados. These forecasts provide local and regional decision-makers with timely, actionable information to mitigate or prevent an outbreak from occurring.”

National dengue early warning system in Barbados

In partnership with the Barbados Ministry of Health and Wellness and other regional health and meteorological agencies, the team applied the model in a real-world context ahead of the ICC Men's Twenty20 (T20) Cricket World Cup in June 2024. The model predicted a 95% probability of a dengue outbreak based on observed cases and forecasted climate conditions as of March 2024, prompting additional checks and re-treatments of known mosquito breeding sites around the cricket venue and nearby communities ahead of the tournament.

ICREA Professor and leader of the Global Health Resilience group, Rachel Lowe, senior author of the study, emphasized the broader significance of the research: “This study brings us a step closer to implementing early warning systems for climate-sensitive infectious diseases in the Caribbean. What we have piloted in Barbados could be adapted to other regions facing similar public health risks.” From 2025, the model will become part of a national dengue early warning system in Barbados.

Looking ahead, the research team plans to validate further the model and explore its application to other climate-sensitive diseases and geographic contexts. Their work offers a powerful example of how interdisciplinary collaboration—spanning epidemiology, climate science, and public health—can help mitigate disease risks in an era of increasing climatic uncertainty.

Caption attached picture: A Campbell-Stokes sunshine recorder which measures daily sunlight duration located at the Barbados Meteorological Service weather station, Charnocks, Barbados. Credit: Rachel Lowe (BSC).

BSC media contact: communication@bsc.es