key: cord-349645-6o8773c5
authors: Li, He; Xu, Xiao-Long; Dai, Da-Wei; Huang, Zhen-Yu; Ma, Zhuang; Guan, Yan-Jun
title: Air Pollution and temperature are associated with increased COVID-19 incidence: a time series study
date: 2020-06-02
journal: Int J Infect Dis
DOI: 10.1016/j.ijid.2020.05.076
sha: 
doc_id: 349645
cord_uid: 6o8773c5

OBJECTIVES: Although the COVID-19 is known to cause by human-to-human transmission, it remains largely unclear whether ambient air pollutants and meteorological parameters could promote its transmission. METHODS: A retrospective study is conducted to study whether air quality index (AQI), four ambient air pollutants (PM(2.5), PM(10), NO(2) and CO) and five meteorological variables (daily temperature, highest temperature, lowest temperature, temperature difference and sunshine duration) could increase COVID-19 incidence in Wuhan and XiaoGan between Jan 26(th) to Feb 29(th) in 2020. RESULTS: First, a significant correlation was found between COVID-19 incidence and AQI in both Wuhan (R(2) = 0.13, p < 0.05) and XiaoGan (R(2) = 0.223, p < 0.01). Specifically, among four pollutants, COVID-19 incidence was prominently correlated with PM(2.5) and NO(2) in both cities. In Wuhan, the tightest correlation was observed between NO(2) and COVID-19 incidence (R(2) = 0.329, p < 0.01). In XiaoGan, in addition to the PM(2.5) (R(2) = 0.117, p < 0.01) and NO(2) (R(2) = 0.015, p < 0.05), a notable correlation was also observed between the PM(10) and COVID-19 incidence (R(2) = 0.105, p < 0.05). Moreover, temperature is the only meteorological parameter that constantly correlated well with COVID-19 incidence in both Wuhan and XiaoGan, but in an inverse correlation (p < 0.05). CONCLUSIONS: AQI, PM(2.5), NO(2), and temperature are four variables that could promote the sustained transmission of COVID-19.

The outbreak of COVID-19 from Wuhan China had been officially characterized as a pandemic on 11 th March, 2020 (World Health Organization, 2020b , which has led to more than 3.5 million subjected infected and 0.24 million dead worldwide as of May 4 th 2020 (World Health Organization, 2020a) . The causative pathogen of COVID-19 has been confirmed as the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) (Lu et al., 2020) , which belongs to the coronavirus family and previously caused severe acute respiratory syndrome (SARS) (Peiris et al., 2004) and the Middle East respiratory syndrome (MERS) (Zaki et al., 2012) .

Air pollution has been an on-going research focus as it is a major environmental threat to human health. Sufficient evidence has tightly linked ambient air pollution to occurrence of numerous respiratory disease, such as COPD (Ling and van Eeden, 2009 ) and asthma (Gorai et al., 2016) . Moreover, air pollution is also associated with infectious diseases transmission. For example, worsen air quality has also been shown to J o u r n a l P r e -p r o o f increase SARS fatality (Cui et al., 2003) as well as increased influenza incidence (Landguth et al., 2020) . In laboratory condition, van Doremalen (van Doremalen et al., 2020) has demonstrated a long viability of SARS-CoV-2 in ambient aerosols, which could be an important source of COVID-19 transmission (Luo et al., 2020) . However, whether ambient air pollutants are associated with increased incidence of COVID-19 in the realistic situation remains largely unknown.

Previous study indicated that meteorological parameters can affect spread and thrive of multiple virus. For example, ambient temperature and relative humidity are inversely associated with influenza A infection rate in Japan (Iha et al., 2016) . The coronavirus also exhibited a seasonal oscillation of outbreak, which also suggested a strong association between meteorological parameters and virus transmission and viability (Killerby et al., 2018) . Moreover, although the epidemiological characteristics of SARS-CoV-2 is not clear, a recent study predicted SARS-CoV-2 transmits more efficiently in winter than summer (Lipsitch, 2020) , indicating the importance of temperature in COVID-19 transmission. However, whether COVID-19 transmission is associated with meteorological parameters, at this moment, is not backed by sufficient investigations and robust evidence.

In this retrospective study, we attempted to conduct an exploratory analysis looking at the association between environment conditions (including ambient pollutants and meteoroidal parameter) and COVID-19 incidence/mortality in Wuhan, given a city-wide lockdown and varying pollution/meteorological data throughout the entire study period.

In this time-series analysis, COVID-19 incidence counts in Wuhan and XiaoGan were provided by the Centers for Disease Control and Prevention (CDC) of Hubei Province (Health Commission of the Hubei Province, 2020). The data used in this study are daily case counts of positive J o u r n a l P r e -p r o o f diagnoses of COVID-19 from all reporting sources, including laboratory and clinical diagnoses. COVID-19 cases of all ages are included. In total, the COVID-19 incidence data for Wuhan produced over 35 'clusters' of time series between Jan 26 th to Feb 29 th in 2020.

The daily air quality index (AQI), PM 2.5 , PM 10 , NO 2 and CO concentration were retrieved from the Platform AQI (Platform AQI, 2020).

Five meteorological parameters were retrieved from the database of Weather.com (The Weather Channel, 2020), including daily mean temperature, highest temperature, lowest temperature, sun-rise and sun-set time. Daily temperature difference and sunshine duration were calculated based on the difference of highest and lowest temperature and sun rise and set time, respectively. Thus, a total of five meteorological parameters were enrolled as independent variables.

Due to imperfect daily reporting practices, COVID-19 incidence number in XiaoGan exhibited a data on Feb 19 th , which was -15.

Furthermore, China updated their diagnostic criteria on Feb 12 th and 13 th , which resulted in a significant increase of COVID-19 incidence cases on these days (Han and Yang, 2020) . Thus, these data were excluded from the current study.

All data analyses were used in GraphPad Prism ® 8.0 (GraphPad Software, La Jolla California USA). First, a descriptive analysis were performed to provide an overview of COVID-19 incidence and air quality during the study period. Next, we utilized linear regression model to fit the dependent variables (COVID-19 incidence) for each independent variable (four ambient air pollutants and five meteorological parameters). As the SARS-CoV-2 has a median incubation period of 4 days in human (Guan et al., 2020) , all independent variables were used to fit daily COVID-19 incidence from 4 days later. The statistical tests were two-sided, and p-value <0.05 was considered as statistically significant. Table 1 , including the AQI, four ambient air pollutants and five meteorological parameters. The AQI value for Wuhan and XiaoGan were 63.63 + 4.47 and 71.58 + 5, respectively, both of which were categorized as moderate according to US EPA standard (United States Environmental Protection Agency, 2019a). The highest and lowest daily COVID-19 incidence were 1,690 cases on Feb 16 th and 80 cases on Jan 26 th in Wuhan, respectively. Meanwhile, that of XiaoGan were 424 on Feb 5 th and 0 on Feb 29 th , respectively.

We then looked into the correlation between local AQI and COVID-19 incidence number in each city ( Table 2 , Fig 1) . The data showed that AQI was significantly and positively associated with daily COVID-19 incidence number in both Wuhan (R 2 = 0.13, p<0.05) and XiaoGan (R 2 = 0.223, p<0.01), which indicating the important role of AQI in COVID-19 transmission. Thus, we further studied the association of daily newly diagnosed COVID-19 cases with each air pollutant in Wuhan (Table 2, Fig 2) . Interestingly, all ambient air pollutants showed positive association with daily COVID-19 incidence. Among them, NO 2 (R 2 = 0.329, p<0.01), PM 2.5 (R 2 = 0.174, p<0.05) and CO (R 2 = 0.203, p<0.001) exhibited a statistical significance. Next, we studied the correlation between meteorological parameters and COVID-19 incidence in Wuhan. Among five parameters, daily temperature (R 2 = 0.126, p<0.05) and daily lowest temperature (R 2 = 0.143, p<0.05) were predominantly correlated with COVID-19 incidence, but both in an inverse correlation.

In XiaoGan, both PM 2.5 (R 2 = 0.23, p<0.01) and NO 2 (R 2 = 0.158, p<0.05) were also apparently associated with COVID-19 incidence.

Moreover, a notable correlation was also observed between the PM 10 and incidence cases (R 2 = 0.158, p<0.05). Among five meteorological factors, COVID-19 incidence correlated well with the temperature (R 2 = 0.13, p<0.05) and daily sunshine duration (R 2 = 0.407, p<0.01), which were, however, in an inverse correlation.

Multiple factors could impact viral transmission. For example, influenza viability and activity could be potentiated by ambient air pollutants and some meteorological variables (Iha et al., 2016 , Landguth et al., 2020 . However, this has not been examined for the SARS-CoV-2. Thus, a preliminary analysis was conducted in the current study to assess the role of air pollution and meteorological parameters on COVID-19 transmission.

We found that COVID-19 incidence was enhanced by increased AQI (decreased air quality), PM 2.5 , and NO2 and weakened by temperature.

The PM is hazardous due to its complicate composition and strong capacity of air suspension. PM could be divided into coarse and fine particulate matter, whose diameter are less than 10 μm and 2.5 μm, respectively. Among various causes of respiratory illness, the PM has been shown to potentiate viral transmissions. For instance, ambient PM 2.5 concentration was prominently correlated with influenza-like illness risk at the flu season in Beijing, China (Feng et al., 2016) . In a single hospital setting, researchers found that both influenza and respiratory syncytial virus remain airborne for a long time period after they attach to PM, which allows viruses to be transmitted by the airborne route (Lindsley et al., 2010) .

Additionally, viral replication in the respiratory system is enhanced by the negative effect of PM on the human respiratory barrier integrity (Xian et al., 2020 , Zhao et al., 2018 . However, it remains particularly unclear whether the ambient air pollutant could assist SARS-CoV-2 transmission. In the current study, the PM 2.5 concentration is correlated with the COVID-19 incidence in a positive correlation, which agreed with previous studies.

Based on these information, we hypothesize that there are PM could potentiate the transmission ability of SARS-CoV-2 in two ways: 1) PM 2.5 could disrupt the integrity of human respiratory barrier integrity (Zhao et al., 2018) . Thus, the dysfunctional respiratory barrier are more likely to expose deeper respiratory tissue to foreign pathogens. 2) PM could form condensation nuclei for viral attachment (Lee et al., 2014) . Due to its relative smaller size, PM 2.5 are more pernicious as it can penetrate the respiratory tract and reach alveoli directly (Tellier, 2009 ). Since both PM, especially J o u r n a l P r e -p r o o f PM 2.5 concentration, were constantly higher than safety guideline of US EPA (United States Environmental Protection Agency, 2019b) in both cities, we believed that PM 2.5 is a stronger factor promoting SARS-CoV-2 transmission.

There are a number of studies demonstrate the adverse health effects of NO 2 exposure. For example, short-term increase of outdoor NO 2 concentration can significantly increase the risk of upper respiratory tract infection (Li et al., 2018) . This phenomenon is was particularly notable in children, as this subpopulation is highly susceptible to NO 2 induced lung injury (Ghosh et al., 2012 , Lin et al., 2013 , Moshammer et al., 2006 . Viral infection was common after NO 2 exposure. According to Chauhan et al. (Chauhan et al., 2003) , four viruses were frequently detected in NO 2 -related respiratory tract infection and coronavirus was one of them. Previous study indicated that preceding NO 2 exposure can decrease host immunity and thus significantly increase infection risk of cytomegalovirus in mice (Rose et al., 1988) . Moreover, recovered mice tended to be reinfected after re-exposing to NO 2 (Rose et al., 1989 ). In the current study, although the NO 2 level was constantly lower than the US EPA standards (United States Environmental Protection Agency, 2016), our data revealed that COVID-19 incidence were highly correlated with the ambient NO 2 concentration. This finding agreed to epidemiological studies from other regions of the world (Chauhan et al., 2003 , Lin et al., 2013 .

So far, epidemiological studies had identified at least nine viruses' categories that are capable of infecting respiratory tract (Nichols et al., 2008 , Pavia, 2011 . Although all feature seasonal oscillation of outbreaks, only three viruses show peak incidences in the winter months, which are the Influenza, human coronavirus, and human respiratory syncytial virus (Killerby et al., 2018 , Midgley et al., 2007 . Although the epidemiological characteristics of SARS-CoV-2 is not clear, recent study predicted the SARS-CoV-2 transmits more efficiently in winter than summer (Lipsitch, 2020) , indicating the importance of temperature. These data agreed with our results as temperature seems to decrease the incidence of COVID-19, indicating an inhibitory effect of temperature on SARS-CoV-2 transmission. This phenomenon might be related to life-style as people tend to huddle indoors together during winter season. Future study needs to investigate the direct effect of temperature on viral activity as well.

Other than the PM and NO 2 , the data from Wuhan also indicated that CO have a strong positive effect on SARS-CoV-2 transmission.

However, the data from XiaoGan failed to repeat the same result. So far, only a few studies are available concerning the effect of CO on viral transmission. For example, Su et al.(Su et al., 2019) presented that CO can increase the risk of influenza-like illness. Ali also identified that CO had a weak positive association with influenza transmissibility (Ali et al., 2018) . We believed our current results could not fully represent the potency of CO on SARS-CoV-2 transmission due to relatively limited study period and location. Thus, we cannot conclude the effect of CO on SARS-CoV-2 transmission based on current data. Further studies are required to elaborate this issue.

The current study has some limitations. First, there are only two cities enrolled, which might result in some results deviation from the exact effect of ambient pollution and meteorological parameters on SARS-CoV-2 transmission. Second, the study period is relative short compared to other epidemiological study. In future study, we will enroll more data from multiple countries and areas to validate the results from current study.

In conclusion, we found that AQI, PM 2.5 , NO 2 and temperature are four variables that could potential promote the sustained transmission of SARS-CoV-2. Personal protective devices, especially the facial mask, shall be suggested to residents for coronavirus protection in highly polluted regions.

These is no conflict of interest in this study. Sunshine Duration -0.000 + 0.000 0.000 -0.002 + 0.000 *** 0.407 Table 2 . The correlation between the COVID-19 incidence and three ambient air pollution along with five meteorological parameters Jan 26 th to Feb 29 th in 2020 in Wuhan and XiaoGan, China. The data marked with *, ** and *** indicated statistical significance, where p <0.05, 0.01 and 0.001, respectively.

Ambient ozone and influenza transmissibility in Hong Kong

Personal exposure to nitrogen dioxide (NO2) and the severity of virus-induced asthma in children

Air pollution and case fatality of SARS in the People's Republic of China: an ecologic study

Impact of ambient fine particulate matter (PM2.5) exposure on the risk of influenza-like-illness: a time-series analysis in Beijing

Hertz-Picciotto I. Ambient nitrogen oxides exposure and early childhood respiratory illnesses

Association between Ambient Air Pollution and Asthma Prevalence in Different Population Groups Residing in Eastern Texas

Clinical Characteristics of Coronavirus Disease 2019 in China

The transmission and diagnosis of 2019 novel coronavirus infection disease (COVID-19): A Chinese perspective

Health Commission of the Hubei Province. The realtime coronovirus epidemiological situationof Hubei province

Comparative epidemiology of influenza A and B viral infection in a subtropical region: a 7-year surveillance in Okinawa

Human coronavirus circulation in the United States

The delayed effect of wildfire season particulate matter on subsequent influenza season in a mountain west region of the USA

Exposure to combustion generated environmentally persistent free radicals enhances severity of influenza virus infection

Association between air pollution and upper respiratory tract infection in hospital outpatients aged 0-14 years in Hefei, China: a time series study

Temperature, nitrogen dioxide, circulating respiratory viruses and acute upper respiratory infections among children in Taipei, Taiwan: a population-based study

Distribution of airborne influenza virus and respiratory syncytial virus in an urgent care medical clinic

Particulate matter air pollution exposure: role in the development and exacerbation of chronic obstructive pulmonary disease

Seasonality of SARS-CoV-2: Will COVID-19 go away on its own in warmer weather?

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding

Possible Transmission of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) in a Public

Determining the seasonality of respiratory syncytial virus in the United States: the impact of Increased molecular testing

Low levels of air pollution induce changes of lung function in a panel of schoolchildren

Respiratory viruses other than influenza virus: impact and therapeutic advances

Viral infections of the lower respiratory tract: old viruses, new viruses, and the role of diagnosis

Online air quality monitoring and analysis platform

The pathophysiology of enhanced susceptibility to murine cytomegalovirus respiratory infection during short-term exposure to 5 ppm nitrogen dioxide

Altered susceptibility to viral respiratory infection during short-term exposure to nitrogen dioxide

The short-term effects of air pollutants on influenza-like illness in Jinan, China

Aerosol transmission of influenza A virus: a review of new studies

United States Environmental Protection Agency. What are the Air Quality Standards for PM?

World Health Organization. Coronavirus disease (COVID-19) outbreak situation

Particulate matter 2.5 causes deficiency in barrier integrity in human nasal epithelial cells

Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia

Nasal epithelial barrier disruption by particulate matter ≤2.5 μm via tight junction protein degradation

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

We want to thank all medical workers for their great contribution and sacrifice during this COVID-19 pandemics. We also want to thank the financial support from Natural Science Foundation of Liaoning Province, China (20180551153).J o u r n a l P r e -p r o o f