Rachel Lowe1, Sophie A Lee2, Kathleen M O'Reilly3, Oliver J Brady3, Leonardo Bastos4, Gabriel Carrasco-Escobar5, Rafael de Castro Catão6, Felipe J Colón-González2, Christovam Barcellos7, Marilia Sá Carvalho8, Marta Blangiardo9, Håvard Rue10, Antonio Gasparrini11. 1. Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK; Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK; Department of Infectious Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, UK. Electronic address: rachel.lowe@lshtm.ac.uk. 2. Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK; Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK; Department of Infectious Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, UK. 3. Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK; Department of Infectious Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, UK. 4. Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK; Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK; Department of Infectious Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, UK; Scientific Computing Program, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil. 5. Scripps Institution of Oceanography, University of California San Diego, San Diego, CA, USA; Health Innovation Laboratory, Institute of Tropical Medicine Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru. 6. Department of Geography, Federal University of Espírito Santo, Vitória, Brazil. 7. Health Information and Communication Institute, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil. 8. Scientific Computing Program, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil. 9. MRC Centre for Environment and Health, Department of Epidemiology and Biostatistics, Imperial College London, London, UK. 10. Statistics Program, Computer, Electrical and Mathematical Sciences and Engineering Division, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia. 11. Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK; Department of Public Health Environments and Society, London School of Hygiene & Tropical Medicine, London, UK; Centre for Statistical Modelling, London School of Hygiene & Tropical Medicine, London, UK.
Abstract
BACKGROUND: Temperature and rainfall patterns are known to influence seasonal patterns of dengue transmission. However, the effect of severe drought and extremely wet conditions on the timing and intensity of dengue epidemics is poorly understood. In this study, we aimed to quantify the non-linear and delayed effects of extreme hydrometeorological hazards on dengue risk by level of urbanisation in Brazil using a spatiotemporal model. METHODS: We combined distributed lag non-linear models with a spatiotemporal Bayesian hierarchical model framework to determine the exposure-lag-response association between the relative risk (RR) of dengue and a drought severity index. We fit the model to monthly dengue case data for the 558 microregions of Brazil between January, 2001, and January, 2019, accounting for unobserved confounding factors, spatial autocorrelation, seasonality, and interannual variability. We assessed the variation in RR by level of urbanisation through an interaction between the drought severity index and urbanisation. We also assessed the effect of hydrometeorological hazards on dengue risk in areas with a high frequency of water supply shortages. FINDINGS: The dataset included 12 895 293 dengue cases reported between 2001 and 2019 in Brazil. Overall, the risk of dengue increased between 0-3 months after extremely wet conditions (maximum RR at 1 month lag 1·56 [95% CI 1·41-1·73]) and 3-5 months after drought conditions (maximum RR at 4 months lag 1·43 [1·22-1·67]). Including a linear interaction between the drought severity index and level of urbanisation improved the model fit and showed the risk of dengue was higher in more rural areas than highly urbanised areas during extremely wet conditions (maximum RR 1·77 [1·32-2·37] at 0 months lag vs maximum RR 1·58 [1·39-1·81] at 2 months lag), but higher in highly urbanised areas than rural areas after extreme drought (maximum RR 1·60 [1·33-1·92] vs 1·15 [1·08-1·22], both at 4 months lag). We also found the dengue risk following extreme drought was higher in areas that had a higher frequency of water supply shortages. INTERPRETATION: Wet conditions and extreme drought can increase the risk of dengue with different delays. The risk associated with extremely wet conditions was higher in more rural areas and the risk associated with extreme drought was exacerbated in highly urbanised areas, which have water shortages and intermittent water supply during droughts. These findings have implications for targeting mosquito control activities in poorly serviced urban areas, not only during the wet and warm season, but also during drought periods. FUNDING: Royal Society, Medical Research Council, Wellcome Trust, National Institutes of Health, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, and Conselho Nacional de Desenvolvimento Científico e Tecnológico. TRANSLATION: For the Portuguese translation of the abstract see Supplementary Materials section.
BACKGROUND: Temperature and rainfall patterns are known to influence seasonal patterns of dengue transmission. However, the effect of severe drought and extremely wet conditions on the timing and intensity of dengue epidemics is poorly understood. In this study, we aimed to quantify the non-linear and delayed effects of extreme hydrometeorological hazards on dengue risk by level of urbanisation in Brazil using a spatiotemporal model. METHODS: We combined distributed lag non-linear models with a spatiotemporal Bayesian hierarchical model framework to determine the exposure-lag-response association between the relative risk (RR) of dengue and a drought severity index. We fit the model to monthly dengue case data for the 558 microregions of Brazil between January, 2001, and January, 2019, accounting for unobserved confounding factors, spatial autocorrelation, seasonality, and interannual variability. We assessed the variation in RR by level of urbanisation through an interaction between the drought severity index and urbanisation. We also assessed the effect of hydrometeorological hazards on dengue risk in areas with a high frequency of water supply shortages. FINDINGS: The dataset included 12 895 293 dengue cases reported between 2001 and 2019 in Brazil. Overall, the risk of dengue increased between 0-3 months after extremely wet conditions (maximum RR at 1 month lag 1·56 [95% CI 1·41-1·73]) and 3-5 months after drought conditions (maximum RR at 4 months lag 1·43 [1·22-1·67]). Including a linear interaction between the drought severity index and level of urbanisation improved the model fit and showed the risk of dengue was higher in more rural areas than highly urbanised areas during extremely wet conditions (maximum RR 1·77 [1·32-2·37] at 0 months lag vs maximum RR 1·58 [1·39-1·81] at 2 months lag), but higher in highly urbanised areas than rural areas after extreme drought (maximum RR 1·60 [1·33-1·92] vs 1·15 [1·08-1·22], both at 4 months lag). We also found the dengue risk following extreme drought was higher in areas that had a higher frequency of water supply shortages. INTERPRETATION: Wet conditions and extreme drought can increase the risk of dengue with different delays. The risk associated with extremely wet conditions was higher in more rural areas and the risk associated with extreme drought was exacerbated in highly urbanised areas, which have water shortages and intermittent water supply during droughts. These findings have implications for targeting mosquito control activities in poorly serviced urban areas, not only during the wet and warm season, but also during drought periods. FUNDING: Royal Society, Medical Research Council, Wellcome Trust, National Institutes of Health, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, and Conselho Nacional de Desenvolvimento Científico e Tecnológico. TRANSLATION: For the Portuguese translation of the abstract see Supplementary Materials section.
Authors: Beatrice R Egid; Mamadou Coulibaly; Samuel Kweku Dadzie; Basile Kamgang; Philip J McCall; Luigi Sedda; Kobie Hyacinthe Toe; Anne L Wilson Journal: Curr Res Parasitol Vector Borne Dis Date: 2022
Authors: Yuyang Chen; Naizhe Li; José Lourenço; Lin Wang; Bernard Cazelles; Lu Dong; Bingying Li; Yang Liu; Mark Jit; Nikos I Bosse; Sam Abbott; Raman Velayudhan; Annelies Wilder-Smith; Huaiyu Tian; Oliver J Brady Journal: Lancet Infect Dis Date: 2022-03-02 Impact factor: 71.421
Authors: G Harsha; T S Anish; A Rajaneesh; Megha K Prasad; Ronu Mathew; Pratheesh C Mammen; R S Ajin; Sekhar L Kuriakose Journal: GeoJournal Date: 2022-09-16