Chloe Fletcher

Brazil has one of the highest dengue burdens globally, with transmission influenced by climate events. Yet, the effects of compound climate extremes on dengue risk and prediction have not been quantified to date. Using a Bayesian spatiotemporal model of monthly dengue cases across 435 health regions from 2010 to 2025, we estimated relative risk associated with interacting long-lag and short-lag drought indicators and temperature anomalies across nine Köppen-Geiger climatic zones, adjusting for absolute temperature and El Niño-Southern Oscillation. Long-lag dry, mid-lag warm and short-lag wet conditions were associated with the highest relative risk in five climatic zones, with temperature anomalies modifying hydrometeorological dynamics more strongly in temperate than tropical areas. Climate covariates improved the prediction of case counts and outbreaks compared to baseline models at lead times of 1, 3, 6 and 12 months, and for the upcoming season 5 months ahead. Improvements were greatest in the Northeast, Southeast, and Midwest. Recent dengue incidence provided the greatest improvement at the 1-month lead, whereas combining incidence with climate information performed best at 3- and 6-month leads. We extended our framework to weekly state-level predictions of the season ahead to align with national public health planning. We proposed a multi-staged dengue forecasting approach to enhance early warning and response across Brazil.

Dengue presents a major public health challenge in Vietnam, driven by biological, behavioural, and environmental factors. Compound climate extremes, such as sequential hydrometeorological events, can influence dengue risk yet remain understudied. We evaluated the effects of compound climate extremes, socioeconomic conditions, and human mobility on dengue relative risk across 670 districts in Vietnam over 20 years, stratifying by eight subregions spanning emerging to endemic transmission. Dengue risk was greatest following dry-then-wet conditions in seven subregions. Higher temperatures increased risk across North and Central Vietnam, but had limited effect in the South. Mobility associations followed an urban-rural gradient, with increased risk in highly rural districts where residents travelled more frequently to fewer destinations, and in urban districts with more visitors and dispersed outgoing mobility. Stratifying climatic and socioeconomic effects by subregion improved predictive skill at lead times of 1 to 6 months over unstratified and baseline models, with substantial spatial variation. Accounting for compound extremes across distinct spatial contexts could strengthen disease early warning systems in Vietnam and beyond.

Background: Malaria remains a leading cause of morbidity and mortality in South Sudan, with transmission influenced by variations in temperature and rainfall. Since 2013, Médecins Sans Frontières (MSF) has systematically recorded all confirmed malaria cases from MSF-run health services in Lankien thus enabling the development and testing of a malaria early warning system at operationally relevant lead times for anticipatory humanitarian action.Methods: We developed several Bayesian models to probabilistically forecast malaria outbreak risk 1- to 7-months ahead using climate, environmental, and lagged malaria case data. Models were trained on June 2013-May 2025 data and selected using exploratory analyses and goodness-of-fit metrics and compared against a seasonal baseline model reflecting current outbreak monitoring practice.Findings: Across 1-7-month lead times, climate-informed models outperformed a seasonal-only baseline, although predictive accuracy declined at longer lead times. Key predictors included lagged malaria cases, mean temperature (2-month lag), consecutive dry days (4-month lag), and the 6-month average Standardised Precipitation Evapotranspiration Index (SPEI-6; 8-month lag). Short-term wetter conditions were associated with increased risk and longer-term dry conditions with increased risk.Interpretation: Early warnings issued at different lead times and during different seasons guide distinct operational decisions. Longer-lead forecasts (>3 months) support large-scale preventive interventions, while shorter-lead forecasts enable rapid responses such as vector control or community clean-ups. Pre-rainy season early warnings can inform supply pre-positioning, whereas early warnings during the flood-prone rainy season when logistics are constrained can guide health promotion and surge capacity efforts. Rolling 1–7-month predictions extend preparedness phases and increase confidence when multiple forecasts align.

Forecast models are a key decision-support tool for public health authorities in managing epidemics, feeding into early warning systems, scenario evaluations, and an empirical basis for resource allocation. In Brazil, improving dengue forecasting became a priority in response to the unprecedented increase in cases, which surpassed the total of the previous decade and expanded to new regions. The Infodengue-Mosqlimate consortium launched the Infodengue-Mosqlimate Dengue Challenge 2024 (IMDC24), or Dengue Forecast Sprint, bringing together six international teams provided with cases and climate covariates data to generate actionable forecasts for 2024 and 2025 seasons in five diverse Brazilian states, leveraging advanced machine learning and classical statistical models. This paper outlines the structure and findings of the IMDC24. The performance of the models varied between years and locations, and no single model consistently excelled, especially during 2024’s unprecedentedly large season. This performance variability highlighted the need for ensemble approaches. The ensemble models developed are presented as the main results of this collaborative development. As intended, the ensemble models have been adopted by Brazilian public health authorities to help with planning and response to the forecasted 2025 dengue epidemics across the country.

Forecast models are a key decision-support tool for public health authorities in managing epi- demics, feeding into early warning systems, scenario evaluations, and empirical basis for resource al- location. In Brazil, improving dengue forecasting became a priority in response to the unprecedented increase in cases, which surpassed the total of the previous decade and expanded to new regions. The Infodengue-Mosqlimate consortium launched the Brazilian Dengue 2024 Challenge (IMDC24), or Dengue Forecast Sprint, bringing together six international teams provided with cases and climate covariates data to generate actionable forecasts for 2024 and 2025 seasons in five diverse Brazilian states, leveraging advanced machine learning and classical statistical models. This paper outlines the structure and findings of the IMDC24. Model performance varied between years and locations, and no single model consistently excelled, especially during 2024’s atypical, climate-change-driven con- ditions. This performance variability highlighted the need for ensemble approaches. The ensemble models developed are presented as the main results of this collaborative development. As intended, the ensemble models have been adopted by Brazilian public health authorities to help with planning and response to the forecasted 2025 dengue epidemics across the country. Significance Statement: The Dengue Challenge 2024 (IMDC24) was organized by the Mosqlimate-Infodengue consortium, which aims to provide forecasting models as decision support tools for early warning systems, scenario assess- ments, and empirical basis for resource allocation for mosquito-borne diseases. During IMDC24, six international teams, provided with dengue case, sociodemographic, and climate data, developed scenario forecasting models for the 2024 and 2025 dengue seasons in Brazil. In this study, we evaluated the per- formance of each model and built an ensemble model, considering the variation in performance of each model, especially during the atypical climate conditions of 2024. Among the main applications of this work, we highlight the incorporation of the results into the Brazilian Ministry of Health’s nationwide dengue epidemic response agenda.

Here we introduce a demand-driven framework designed to implement climate services in the health sector, with a particular focus on the Caribbean region. Climate services are essential for supporting informed decision-making and response strategies in relation to climate-related health risks. Through collaborative efforts, we are co-producing a climate-driven dengue early warning system (EWS) to target vector-borne diseases effectively. While challenges exist in implementing such systems, EWSs provide valuable tools for managing epidemic risks by predicting potential disease outbreaks in advance. The scarcity of operational climate tools in the health sector underscores the need for increased investment and strategic implementation practices. To address these challenges, a demand-driven framework is proposed, emphasizing strategic planning focused on health intervention development, partnership building, data, communication, human resources, capacity building, and sustainable funding. This framework aims to integrate climate services seamlessly into health systems, thereby enhancing public health resilience and facilitating well-informed decision-making to effectively address climate-sensitive diseases.