PDF(Pubmed)
Abstract：
There is growing evidence that weather alters SARS-CoV-2 transmission, but it remains unclear what drives the phenomenon. One prevailing hypothesis is that people spend more time indoors in cooler weather, leading to increased spread of SARS-CoV-2 related to time spent in confined spaces and close contact with others. However, the evidence in support of that hypothesis is limited and, at times, conflicting. We use a mediation framework, and combine daily weather, COVID-19 hospital surveillance, cellphone-based mobility data and building footprints to estimate the relationship between daily indoor and outdoor weather conditions, mobility, and COVID-19 hospitalizations. We quantify the direct health impacts of weather on COVID-19 hospitalizations and the indirect effects of weather via time spent indoors away-from-home on COVID-19 hospitalizations within five Colorado counties between March 4th 2020 and January 31st 2021. We also evaluated the evidence for seasonal effect modification by comparing the results of all-season (using season as a covariate) to season-stratified models. Four weather conditions were associated with both time spent indoors away-from-home and 12-day lagged COVID-19 hospital admissions in one or more season: high minimum temperature (all-season), low maximum temperature (spring), low minimum absolute humidity (winter), and high solar radiation (all-season & winter). In our mediation analyses, we found evidence that changes in 12-day lagged hospital admissions were primarily via the direct effects of weather conditions, rather than via indirect effects by which weather changes time spent indoors away-from-home. Our findings do not support the hypothesis that weather impacted SARS-CoV-2 transmission via changes in mobility patterns during the first year of the pandemic. Rather, weather appears to have impacted SARS-CoV-2 transmission primarily via mechanisms other than human movement. We recommend further analysis of this phenomenon to determine whether these findings generalize to current SARS-CoV-2 transmission dynamics, as well as other seasonal respiratory pathogens.
摘要：
越来越多的证据表明，天气改变了SARS-CoV-2的传播，但目前尚不清楚是什么导致了这种现象。一个普遍的假设是，人们在凉爽的天气里呆在室内的时间更多，导致SARS-CoV-2的传播增加，这与在密闭空间中花费的时间以及与他人的密切接触有关。然而,支持该假设的证据是有限的，有时,冲突。我们使用调解框架，结合日常天气，COVID-19医院监测，基于手机的移动数据和建筑足迹，以估计日常室内和室外天气条件之间的关系，移动性,和COVID-19住院。我们量化了天气对COVID-19住院的直接健康影响，以及天气通过在2020年3月4日至2021年1月31日期间在科罗拉多州五个县内离家在室内度过的时间对COVID-19住院的间接影响。我们还通过比较全季节（使用季节作为协变量）与季节分层模型的结果来评估季节效应改变的证据。在一个或多个季节中，四种天气状况与外出在室内度过的时间和12天滞后的COVID-19住院时间有关：较高的最低温度（全季节），低最高温度（弹簧），最低绝对湿度低（冬季），和高太阳辐射（全季节和冬季）。在我们的调解分析中，我们发现有证据表明，12天滞后入院的变化主要是通过天气条件的直接影响，而不是通过间接影响，天气改变了在室内外出的时间。我们的发现不支持以下假设：在大流行的第一年，天气通过流动模式的变化影响了SARS-CoV-2的传播。相反，天气似乎主要通过人类运动以外的机制影响了SARS-CoV-2的传播。我们建议对这种现象进行进一步分析，以确定这些发现是否适用于当前的SARS-CoV-2传播动力学，以及其他季节性呼吸道病原体。
