key: cord-0740677-lviwp8ak
authors: Kotsiou, Ourania S.; Zidros, Thomas; Gourgoulianis, Konstantinos I.
title: Letter to Editor regarding Prata et al. (2020), Temperature significantly changes COVID-19 transmission in (sub)tropical cities of Brazil. Science of Total Environment, v729, 138862
date: 2020-07-28
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2020.141323
sha: 4185535b9de110180669093fd002170e9b74a9fc
doc_id: 740677
cord_uid: lviwp8ak

[Figure: see text]

We read with great interest the well-presented article titled "Temperature significantly changes COVID-19 transmission in (sub)tropical cities of Brazil" by Prata et al. (2020) , analyzing the temperature data sampled from 27 cities in Brazil from February 27 th to April 1 st , 2020 found that temperature had a negative linear relationship with the number of confirmed coronavirus disease 2019 cases. A broadly similar point has also recently been made by other studies (Tosepu et al., 2020; Xie et al., 2020) . However, this notion is judged as Brazil recently jumped to the second place of COVID-19 cases in the world, although the high temperatures.

On the other hand, after rejected lockdown measures, hard-hit Brazil removes data amid rising death toll (Prazeres, 2020) . In that context, there are several caveats to interpreting high incidence and mortality rates in Brazil.

Additionally, to the study of Prata et al. (2020) , we aimed to assess the impact of multiple climatological parameters including daily maximum, average and minimum temperature, dew point temperature, humidity, wind speed and sea-level pressure values on daily and cumulative COVID-19 incidence and mortality in 27 states of Brazil from February 26 th , 2020 to May 31 st , 2020, according to data collected before the "statistical coup" by the Brazilian government on 5 th June 2020 (Prazeres, 2020) J o u r n a l P r e -p r o o f 3 the Weather Underground website, (2020), providing local and long-range weather conditions. The population density was calculated by the number of individuals per unit area (Wikipedia, 2020) . Bivariate analyses of correlations of numerical data between groups were tested by Pearson's R Correlation. Receiver operating curve (ROC) analysis was used to determine the best cutoff point for temperature by measuring the area under the curve (AUC). Stepwise multiple linear regression was used to explain the relationship between average daily or cumulative COVID-19 cases or deaths and meteorological variables. A result was considered statistically significant when the P-value was <0.05. Data were analyzed using IBM SPSS Statistics 23.

The most important linear relationships between meteorological conditions and COVID-19 daily and cumulative incidence or mortality, as were indicated by a correlation coefficient, are presented in Table 1 . We found that daily maximum temperature was weakly inversely correlated with daily confirmed new or cumulative COVID-19 cases and daily confirmed new or cumulative COVID-19 deaths. No linear correlation or a very weak linear correlation were found between average humidity and daily or cumulative COVID-19 incidence or mortality. A very weak or weak negative correlations were found between dew point temperature and daily or cumulative COVID-19 incidence or mortality. A very weak positive correlation was found between average wind speed and daily or cumulative COVID-19 incidence or mortality. Average sea-level atmospheric pressure was weakly positively correlated with daily and cumulative COVID-19 incidence or mortality. A multiple stepwise regression analysis was subsequently performed to assess the effects of multiple factors (daily maximum, average and minimum temperature, dew point temperature, humidity, wind speed, sea-level pressure, and density) on daily and cumulative COVID-19 incidence and mortality (dependent variables). The effects of maximum temperature and dew point temperature, density, average wind Journal Pre-proof J o u r n a l P r e -p r o o f 5 speed, and sea-level pressure that were finally used as the independent variables in the prediction of the new daily confirmed or cumulative COVID-19 cases or deaths were statistically significant (p< 0.05); however, these independent predictors explained only 10-20% of the total variance in the regression models (R 2 =10-20%).

Our findings should be put into context with other critical factors such as the seasons change from summer to winter, the testing delays, the differences in quality of living conditions, and the prevention policies taken by the Governments and death data. Undoubtedly, the facts showed that the bad policy had no temperature and could be more dangerous than a virus. Further experimental and epidemiological studies are urgently needed to evaluate the ongoing COVID-19 pandemic, which constitutes a multifactorial problem requiring multifactorial responses.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors 

Impact of delays on effectiveness of contact tracing strategies for COVID-19: a modelling study. The Lancet Public Health

The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application

Governo esconde totais de mortes e casos da Covid-19 e tira site do ar

Temperature significantly changes COVID-19 transmission in (sub)tropical cities of Brazil

COVID-19 in Brazil: "So what

Correlation between weather and Covid-19 pandemic in Jakarta

Weather Underground

List of Brazilian states by population density

Association between ambient temperature and COVID-19 infection in 122 cities from China