秘鲁消除疟疾环境下天气与疟疾之间的关联：分布滞后分析
Associations between weather and malaria in an elimination setting in Peru: a distributed lag analysis.
作者信息
Heitmann Gabriella Barratt, Wu Xue, Nguyen Anna T, Altamirano-Quiroz Astrid, Fine Sydney, Fernandez-Camacho Bryan, Barja Antony, Cava Renato, Soto-Calle Verónica, Rodriguez Hugo, Carrasco-Escobar Gabriel, Bennett Adam, Llanos-Cuentas Alejandro, Mordecai Erin A, Hsiang Michelle S, Benjamin-Chung Jade
机构信息
Department of Epidemiology and Population Health, School of Medicine, Stanford University, Stanford, CA, USA.
Malaria Elimination Initiative, Institute for Global Health Sciences, University of California San Francisco (UCSF), San Francisco, CA, USA.
出版信息
medRxiv. 2024 Nov 28:2024.11.26.24318000. doi: 10.1101/2024.11.26.24318000.
BACKGROUND
() is the predominant malaria species in countries approaching elimination. In the context of climate change, understanding environmental drivers of transmission can guide interventions, yet evidence is limited, particularly in Latin America.
OBJECTIVES
We estimated the association between temperature and precipitation and malaria incidence in a malaria elimination setting in Peru.
METHODS
We analyzed malaria incidence data from 2021-2023 from 30 communities in Loreto, Peru with hourly weather data from the ERA5 dataset and land cover data from MapBiomas. Predictors included average weekly minimum and maximum temperature, high heat (>90th percentile mean temperature), total weekly precipitation, and heavy rain (>90th percentile total precipitation). We fit non-linear distributed lag models for continuous weather predictors and generalized additive models for binary predictors and the lookback period was 2-16 weeks. Temperature models adjusted for total precipitation; precipitation models adjusted for maximum temperature. We performed subgroup analyses by season, community type, and distance to forest edge.
RESULTS
The median vs. lowest values of weekly average minimum temperature was associated with 2.16 to 3.93-fold higher incidence 3-16 weeks later (5-week lag incidence ratio (IR) =3.93 [95% CI 2.18, 7.09]); for maximum temperature, the association was hump-shaped across lags, with protective associations at 1-2 and 15-16 week lags and 1.07-1.66-fold higher incidence at 6-13 week lags. High heat (>27.5°C) was associated with 1.23 to 1.37-fold higher incidence at 5--9 week lags (9-week lag IR = 1.25 [1.02, 1.53]). Associations between total precipitation and malaria incidence were hump-shaped across lags, with the strongest positive association at 750 mm of precipitation at a 9-week lag (IR=1.56; [1.27, 1.65]). Heavy rain (>186mm) was associated with 1.22-1.60-fold higher incidence at 2-10 week lags (9-week lag IR=1.23 [1.02, 1.49]).
DISCUSSION
Higher temperatures and precipitation were generally associated with higher malaria incidence over 1-4 months.
背景
（）是接近疟疾消除阶段国家的主要疟原虫种类。在气候变化背景下，了解传播的环境驱动因素可指导干预措施，但相关证据有限，尤其是在拉丁美洲。
目的
我们估计了秘鲁疟疾消除地区温度、降水与疟疾发病率之间的关联。
方法
我们分析了秘鲁洛雷托30个社区2021 - 2023年的疟疾发病率数据，以及来自ERA5数据集的每小时气象数据和MapBiomas的土地覆盖数据。预测因素包括每周平均最低和最高温度、高温（>第90百分位数平均温度）、每周总降水量以及大雨（>第90百分位数总降水量）。我们对连续气象预测因素拟合非线性分布滞后模型，对二元预测因素拟合广义相加模型，回顾期为2 - 16周。温度模型对总降水量进行了调整；降水模型对最高温度进行了调整。我们按季节、社区类型和距森林边缘的距离进行了亚组分析。
结果
每周平均最低温度的中位数与最低值相比，在3 - 16周后发病率高2.16至3.93倍（5周滞后发病率比（IR）=3.93 [95%置信区间2.18, 7.09]）；对于最高温度，各滞后时间的关联呈驼峰状，在1 - 2周和15 - 16周滞后时有保护作用，在6 - 13周滞后时发病率高1.07 - 1.66倍。高温（>27.5°C）在5 - 9周滞后时发病率高1.23至1.37倍（9周滞后IR = 1.25 [1.02, 1.53]）。总降水量与疟疾发病率之间的关联在各滞后时间呈驼峰状，在9周滞后降水量为750毫米时正相关最强（IR = 1.56；[1.27, 1.65]）。大雨（>186毫米）在2 - 10周滞后时发病率高1.22 - 1.60倍（9周滞后IR = 1.23 [1.02, 1.49]）。
讨论
在1 - 4个月内，较高的温度和降水通常与较高的疟疾发病率相关。
Zotero 插件