Climate Variability and Dengue Hemorrhagic Fever in Hanoi, Viet Nam, During 2008 to 2015

Forecasting dengue in Bangladesh using meteorological variables with a novel feature selection approach

Dengue, a mosquito-borne viral disease, continues to pose severe risks to public health and economic stability in tropical and subtropical regions, particularly in developing nations like Bangladesh. The necessity for advanced forecasting mechanisms has never been more critical to enhance the effectiveness of vector control strategies and resource allocations. This study formulates a dynamic data pipeline to forecast dengue incidence based on 13 meteorological variables using a suite of state-of-the-art machine learning models and custom features engineering, achieving an accuracy of 84.02%, marking a substantial improvement over existing studies. A novel wrapper feature selection algorithm employing a custom objective function is proposed in this study, which significantly improves model accuracy by 12.63% and reduces the mean absolute percentage error by 70.82%. The custom objective function's output can be transformed to quantify the contribution of each variable to the target variable's variability, providing deeper insights into the workings of black box models. The study concludes that relative humidity is redundant in predicting dengue infection, while meteorological factors exhibit more significant short-term impacts compared to long-term and immediate impacts. Sunshine (hours) emerges as the meteorological factor with the most immediate impact, whereas precipitation is the most impactful predictor over both short-term (8-month lag) and long-term (26-30-month lag) periods. Forecasts for 2024 using the best-performing model predict a rise in dengue cases starting in May, peaking at 24,000 cases per month by August and persisting at high levels through October before declining to half by year-end. These findings offer critical insights into temporal climate effects on dengue transmission, aiding the development of effective forecasting systems.

The 2023 report of the MJA-Lancet Countdown on health and climate change: sustainability needed in Australia's health care sector

The MJA-Lancet Countdown on health and climate change in Australia was established in 2017 and produced its first national assessment in 2018 and annual updates in 2019, 2020, 2021 and 2022. It examines five broad domains: health hazards, exposures and impacts; adaptation, planning and resilience for health; mitigation actions and health co-benefits; economics and finance; and public and political engagement. In this, the sixth report of the MJA-Lancet Countdown, we track progress on an extensive suite of indicators across these five domains, accessing and presenting the latest data and further refining and developing our analyses. Our results highlight the health and economic costs of inaction on health and climate change. A series of major flood events across the four eastern states of Australia in 2022 was the main contributor to insured losses from climate-related catastrophes of $7.168 billion - the highest amount on record. The floods also directly caused 23 deaths and resulted in the displacement of tens of thousands of people. High red meat and processed meat consumption and insufficient consumption of fruit and vegetables accounted for about half of the 87 166 diet-related deaths in Australia in 2021. Correction of this imbalance would both save lives and reduce the heavy carbon footprint associated with meat production. We find signs of progress on health and climate change. Importantly, the Australian Government released Australia's first National Health and Climate Strategy, and the Government of Western Australia is preparing a Health Sector Adaptation Plan. We also find increasing action on, and engagement with, health and climate change at a community level, with the number of electric vehicle sales almost doubling in 2022 compared with 2021, and with a 65% increase in coverage of health and climate change in the media in 2022 compared with 2021. Overall, the urgency of substantial enhancements in Australia's mitigation and adaptation responses to the enormous health and climate change challenge cannot be overstated. Australia's energy system, and its health care sector, currently emit an unreasonable and unjust proportion of greenhouse gases into the atmosphere. As the Lancet Countdown enters its second and most critical phase in the leadup to 2030, the depth and breadth of our assessment of health and climate change will be augmented to increasingly examine Australia in its regional context, and to better measure and track key issues in Australia such as mental health and Aboriginal and Torres Strait Islander health and wellbeing.

Climate change and human health in Vietnam: a systematic review and additional analyses on current impacts, future risk, and adaptation

This study aims to investigate climate change's impact on health and adaptation in Vietnam through a systematic review and additional analyses of heat exposure, heat vulnerability, awareness and engagement, and projected health costs. Out of 127 reviewed studies, findings indicated the wider spread of infectious diseases, and increased mortality and hospitalisation risks associated with extreme heat, droughts, and floods. However, there are few studies addressing health cost, awareness, engagement, adaptation, and policy. Additional analyses showed rising heatwave exposure across Vietnam and global above-average vulnerability to heat. By 2050, climate change is projected to cost up to USD1-3B in healthcare costs, USD3-20B in premature deaths, and USD6-23B in work loss. Despite increased media focus on climate and health, a gap between public and government publications highlighted the need for more governmental engagement. Vietnam's climate policies have faced implementation challenges, including top-down approaches, lack of cooperation, low adaptive capacity, and limited resources.

Association between Climate Factors and Dengue Fever in Asuncion, Paraguay: A Generalized Additive Model

Dengue fever has been endemic in Paraguay since 2009 and is a major cause of public-health-management-related burdens. However, Paraguay still lacks information on the association between climate factors and dengue fever. We aimed to investigate the association between climatic factors and dengue fever in Asuncion. Cumulative dengue cases from January 2014 to December 2020 were extracted weekly, and new cases and incidence rates of dengue fever were calculated. Climate factor data were aggregated weekly, associations between dengue cases and climate factors were analyzed, and variables were selected to construct our model. A generalized additive model was used, and the best model was selected based on Akaike information criteria. Piecewise regression analyses were performed for non-linear climate factors. Wind and relative humidity were negatively associated with dengue cases, and minimum temperature was positively associated with dengue cases when the temperature was less than 21.3 °C and negatively associated with dengue when greater than 21.3 °C. Additional studies on dengue fever in Asuncion and other cities are needed to better understand dengue fever.

Deep learning models for forecasting dengue fever based on climate data in Vietnam

Background

Dengue fever (DF) represents a significant health burden in Vietnam, which is forecast to worsen under climate change. The development of an early-warning system for DF has been selected as a prioritised health adaptation measure to climate change in Vietnam.

Objective

This study aimed to develop an accurate DF prediction model in Vietnam using a wide range of meteorological factors as inputs to inform public health responses for outbreak prevention in the context of future climate change.

Methods

Convolutional neural network (CNN), Transformer, long short-term memory (LSTM), and attention-enhanced LSTM (LSTM-ATT) models were compared with traditional machine learning models on weather-based DF forecasting. Models were developed using lagged DF incidence and meteorological variables (measures of temperature, humidity, rainfall, evaporation, and sunshine hours) as inputs for 20 provinces throughout Vietnam. Data from 1997–2013 were used to train models, which were then evaluated using data from 2014–2016 by Root Mean Square Error (RMSE) and Mean Absolute Error (MAE).

Results and discussion

LSTM-ATT displayed the highest performance, scoring average places of 1.60 for RMSE-based ranking and 1.95 for MAE-based ranking. Notably, it was able to forecast DF incidence better than LSTM in 13 or 14 out of 20 provinces for MAE or RMSE, respectively. Moreover, LSTM-ATT was able to accurately predict DF incidence and outbreak months up to 3 months ahead, though performance dropped slightly compared to short-term forecasts. To the best of our knowledge, this is the first time deep learning methods have been employed for the prediction of both long- and short-term DF incidence and outbreaks in Vietnam using unique, rich meteorological features.

Conclusion

This study demonstrates the usefulness of deep learning models for meteorological factor-based DF forecasting. LSTM-ATT should be further explored for mitigation strategies against DF and other climate-sensitive diseases in the coming years.

Dengue fever (DF) represents a significant health burden worldwide and in Vietnam, which is forecast to worsen under climate change. The development of an early-warning system for DF has been selected as a prioritised health adaptation measure to climate change in Vietnam. This study aimed to use deep learning models to develop a prediction model of DF rates in Vietnam using a wide range of climate factors as input variables to inform public health responses for outbreak prevention in the context of future climate change. The study found that LSTM-ATT outperformed competing models, scoring average places of 1.60 for RMSE-based ranking and 1.90 for MAE-based ranking. Notably, it was able to forecast DF incidence better than LSTM in 12 or 14 out of 20 provinces for MAE or RMSE, respectively. Moreover, LSTM-ATT was able to accurately predict DF incidence and outbreaks up to 3 months ahead, though performance dropped slightly compared to short-term forecasts. This is the first time deep learning methods have been employed for the prediction of both long- and short-term DF incidence and outbreaks in Vietnam using unique, rich climate features, and it demonstrates the usefulness of deep learning models for climate-based DF forecasting.

Geographical heterogeneity and socio-ecological risk profiles of dengue in Jakarta, Indonesia

The aim of this study was to assess the role of climate variability on the incidence of dengue fever (DF), an endemic arboviral infection existing in Jakarta, Indonesia. The work carried out included analysis of the spatial distribution of confirmed DF cases from January 2007 to December 2018 characterising the sociodemographical and ecological factors in DF high-risk areas. Spearman's rank correlation was used to examine the relationship between DF incidence and climatic factors. Spatial clustering and hotspots of DF were examined using global Moran's I statistic and the local indicator for spatial association analysis. Classification and regression tree (CART) analysis was performed to compare and identify demographical and socio-ecological characteristics of the identified hotspots and low-risk clusters. The seasonality of DF incidence was correlated with precipitation (r=0.254, P<0.01), humidity (r=0.340, P<0.01), dipole mode index (r= -0.459, P<0.01) and Tmin (r= -0.181, P<0.05). DF incidence was spatially clustered at the village level (I=0.294, P<0.001) and 22 hotspots were identified with a concentration in the central and eastern parts of Jakarta. CART analysis showed that age and occupation were the most important factors explaining DF clustering. Areaspecific and population-targeted interventions are needed to improve the situation among those living in the identified DF high-risk areas in Jakarta.

Heatwaves and dengue outbreaks in Hanoi, Vietnam: New evidence on early warning

Background

Many studies have shown associations between rising temperatures, El Niño events and dengue incidence, but the effect of sustained periods of extreme high temperatures (i.e., heatwaves) on dengue outbreaks has not yet been investigated. This study aimed to compare the short-term temperature-dengue associations during different dengue outbreak periods, estimate the dengue cases attributable to temperature, and ascertain if there was an association between heatwaves and dengue outbreaks in Hanoi, Vietnam.

Methodology/Principal findings

Dengue outbreaks were assigned to one of three categories (small, medium and large) based on the 50th, 75th, and 90th percentiles of distribution of weekly dengue cases during 2008–2016. Using a generalised linear regression model with a negative binomial link that controlled for temporal trends, temperature variation, rainfall and population size over time, we examined and compared associations between weekly average temperature and weekly dengue incidence for different outbreak categories. The same model using weeks with or without heatwaves as binary variables was applied to examine the potential effects of extreme heatwaves, defined as seven or more days with temperatures above the 95th percentile of daily temperature distribution during the study period. This study included 55,801 dengue cases, with an average of 119 (range: 0 to 1454) cases per week. The exposure-response relationship between temperature and dengue risk was non-linear and differed with dengue category. After considering the delayed effects of temperature (one week lag), we estimated that 4.6%, 11.6%, and 21.9% of incident cases during small, medium, and large outbreaks were attributable to temperature. We found evidence of an association between heatwaves and dengue outbreaks, with longer delayed effects on large outbreaks (around 14 weeks later) than small and medium outbreaks (4 to 9 weeks later). Compared with non-heatwave years, dengue outbreaks (i.e., small, moderate and large outbreaks combined) in heatwave years had higher weekly number of dengue cases (p<0.05). Findings were robust under different sensitivity analyses.

Conclusions

The short-term association between temperature and dengue risk varied by the level of outbreaks and temperature seems more likely affect large outbreaks. Moreover, heatwaves may delay the timing and increase the magnitude of dengue outbreaks.

Dengue fever is one of the most common mosquito-borne viral diseases. Weather extremes such as El Niño event and extreme hot summer can affect dengue incidence rate and dengue outbreaks. More frequent, more intensive and longer lasting heatwaves in the 21st century is anticipated because of global warming, making it necessary to investigate the association between heatwaves and dengue outbreaks. In this study, we estimated 4.6%, 11.6%, and 21.9% of incident dengue cases during small, medium, and large outbreaks attributable to temperature in Hanoi, Vietnam. We also found evidence of an association between heatwaves and dengue outbreaks, with longer delayed effects on large outbreaks than small and medium outbreaks. Compared with non-heatwave years, dengue outbreaks in heatwave years had higher number of dengue cases. Heatwave weather may represent an emerging risk factor or predicator of dengue outbreaks in tropical regions. Future dengue prediction models incorporating heatwaves may help increase the accuracy of predictability.

Epidemiological and Clinical Features of Dengue Infection in Adults in the 2017 Outbreak in Vietnam

Purpose

The clinical features and laboratory results of dengue-infected adult patients admitted to the hospital during the 2017 outbreak were analyzed in this study.

Method

This is a cross-sectional study. 2922 patients aged 18 years or more with dengue fever in National Hospital for Tropical Diseases (NHTD) in the North and Hospital for Tropical Disease (HTD) in the South of Vietnam were recruited in this study.

Result

Patients were admitted in the hospital around the year and concentrated from August to December, in 53/63 (84.0%) provinces in Vietnam, and patients in all ages were affected. The number of patients with dengue fever was 1675 (57.3%), dengue with warning signs 914 (31.3%), and severe dengue 333 (11.4%), respectively. Among patients with severe dengue, severe plasma leakage and dengue shock account for 238 (8.1%), severe organ impairment 73 (2.5%), and severe bleeding 22 (0.75%). The rate of mortality was 0.8%, and the outcome of dengue patients is worse in the elderly and people with underlying diseases.

Conclusion

The 2017 dengue outbreak occurred in a larger scale than in the previous years in terms of time, location, and number of patients. More elderly patients were infected by dengue in this outbreak, and this may contribute to the mortality rate. Clinical manifestations of dengue patients in Southern Vietnam are more typical than the northern, but the rate of severe dengue is not different. The mortality risk and underlying conditions associated with dengue-infected elderly patients are worthy of further investigations in the future.