Air transportation, population density and temperature predict the spread of COVID-19 in Brazil

PeerJ. 2020 Jun 3;8:e9322. doi: 10.7717/peerj.9322. eCollection 2020.

Abstract

There is evidence that COVID-19, the disease caused by the betacoronavirus SARS-CoV-2, is sensitive to environmental conditions. However, such conditions often correlate with demographic and socioeconomic factors at larger spatial extents, which could confound this inference. We evaluated the effect of meteorological conditions (temperature, solar radiation, air humidity and precipitation) on 292 daily records of cumulative number of confirmed COVID-19 cases across the 27 Brazilian capital cities during the 1st month of the outbreak, while controlling for an indicator of the number of tests, the number of arriving flights, population density, proportion of elderly people and average income. Apart from increasing with time, the number of confirmed cases was mainly related to the number of arriving flights and population density, increasing with both factors. However, after accounting for these effects, the disease was shown to be temperature sensitive: there were more cases in colder cities and days, and cases accumulated faster at lower temperatures. Our best estimate indicates that a 1 °C increase in temperature has been associated with a decrease in confirmed cases of 8%. The quality of the data and unknowns limit the analysis, but the study reveals an urgent need to understand more about the environmental sensitivity of the disease to predict demands on health services in different regions and seasons.

Keywords: Climate; Coronavirus; Health; Pandemic.

Grant support

This work was supported by Ministry of Science, Technology, Innovation and Communication (MCTIC) of Brazil, the Brazilian Agency of Higher Education (CAPES) and the Brazilian National Council for Scientific and Technological Development (CNPq). Pedro Pequeno received a postdoctoral fellowship from the Brazilian Agency of Higher Education (CAPES). Clarissa Rosa and Jorge Luiz Souza received a postdoctoral fellowship from the Institutional Training Program of the Brazilian National Council for Scientific and Technological Development (PCI-CNPq). William Magnusson received a productivity grant from CNPq. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.