Please use this identifier to cite or link to this item: http://hdl.handle.net/2445/209392
Title: Global short-term mortality risk and burden associated with tropical cyclones from 1980 to 2019: a multi-country time-series study
Author: Huang, Wenzhong
Li, Shanshan
Vogt, Thomas
Xu, Rongbin
Tong, Shilu
Molina, Tomàs, 1963-
Masselot, Pierre
Gasparrini, Antonio
Armstrong, Ben
Pascal, Mathilde
Royé, Dominic
Fook Sheng Ng, Chris
Vicedo-Cabrera, Ana Maria
Schwartz, Joel
Lavigne, Eric
Kan, Haidong
Goodman, Patrick
Zeka, Ariana
Hashizume, Masahiro
Hurtado Diaz, Magali
De la Cruz Valencia, César
Seposo, Xerxes
Nunes, Baltazar
Madureira, Joana
Kim, Ho
Lee, Whanhee
Tobias, Aurelio
Íñiguez, Carmen
Leon Guo, Yue
Pan, Shih-Chun
Zanobetti, Antonella
Ngoc Dang, Tran
Van Dung, Do
Geiger, Tobias
Otto, Christian
Johnson, Amanda
Hales, Simon
Keywords: Ciclons
Temps (Meteorologia)
Clima tropical
Cyclons
Weather
Tropical climate
Issue Date: Aug-2023
Abstract: Summary Background The global spatiotemporal pattern of mortality risk and burden attributable to tropical cyclones is unclear. We aimed to evaluate the global short-term mortality risk and burden associated with tropical cyclones from 1980 to 2019. Methods The wind speed associated with cyclones from 1980 to 2019 was estimated globally through a parametric wind field model at a grid resolution of 0·5° × 0·5°. A total of 341 locations with daily mortality and temperature data from 14 countries that experienced at least one tropical cyclone day (a day with maximum sustained wind speed associated with cyclones ≥17·5 m/s) during the study period were included. A conditional quasi-Poisson regression with distributed lag non-linear model was applied to assess the tropical cyclone-mortality association. A meta-regression model was fitted to evaluate potential contributing factors and estimate grid cell-specific tropical cyclone effects. Findings Tropical cyclone exposure was associated with an overall 6% (95% CI 4-8) increase in mortality in the first 2 weeks following exposure. Globally, an estimate of 97 430 excess deaths (95% empirical CI [eCI] 71 651-126 438) per decade were observed over the 2 weeks following exposure to tropical cyclones, accounting for 20·7 (95% eCI 15·2-26·9) excess deaths per 100 000 residents (excess death rate) and 3·3 (95% eCI 2·4-4·3) excess deaths per 1000 deaths (excess death ratio) over 1980-2019. The mortality burden exhibited substantial temporal and spatial variation. East Asia and south Asia had the highest number of excess deaths during 1980-2019: 28 744 (95% eCI 16 863-42 188) and 27 267 (21 157-34 058) excess deaths per decade, respectively. In contrast, the regions with the highest excess death ratios and rates were southeast Asia and Latin America and the Caribbean. From 1980-99 to 2000-19, marked increases in tropical cyclone-related excess death numbers were observed globally, especially for Latin America and the Caribbean and south Asia. Grid cell-level and country-level results revealed further heterogeneous spatiotemporal patterns such as the high and increasing tropical cyclone-related mortality burden in Caribbean countries or regions. Interpretation Globally, short-term exposure to tropical cyclones was associated with a significant mortality burden, with highly heterogeneous spatiotemporal patterns. In-depth exploration of tropical cyclone epidemiology for those countries and regions estimated to have the highest and increasing tropical cyclone-related mortality burdens is urgently needed to help inform the development of targeted actions against the increasing adverse health impacts of tropical cyclones under a changing climate.
Note: Reproducció del document publicat a: https://doi.org/10.1016/S2542-5196(23)00143-2
It is part of: 2023, vol. 7, num.8, p. 1-12
URI: http://hdl.handle.net/2445/209392
Related resource: https://doi.org/10.1016/S2542-5196(23)00143-2
ISSN: 2542-5196
Appears in Collections:Articles publicats en revistes (Física Aplicada)

Files in This Item:
File Description SizeFormat 
827590.pdf12.19 MBAdobe PDFView/Open


This item is licensed under a Creative Commons License Creative Commons