Impact of Long-Term Air Pollution on the Case Fatality Rate of COVID-19

Background There is preliminary evidence of the long-term exposure to air pollution will affect the outcome of patients with COVID-19. More information is needed about relationship between long-term exposure to air pollution and case fatality rate (CFR) of patients with COVID-19. Methods In this study, we have collected the data of Air Quality Index (AQI), PM2.5, PM10, SO 2 , NO 2 and O 3 from 14 representative cities in China in the past 5 years, and calculated the case fatality rate of COVID-19 in the corresponding city. First, we explored correlation relationship between CFR and long-term air quality indicators. Then, we try to point out the air pollutants that affect the level of CFR and evaluated their predictive value. Results We have observed a positive correlation between the CFR and AQI (1-year, 3-year, 5-year), PM2.5 (1-year, 3-year, 5-year), and PM10 (1-year, 3-year, 5-year). Meanwhile, AQI (3-year, 5-year) and PM2.5 (1-year, 3-year, 5-year) were signicantly higher in the high CFR group. Moderate predictive value of air pollution indicator to CFR such as AQI (1-year, 3-year, 5-year), PM2.5 (1-year, 3-year, 5-year) have been found. Conclusions Our results indicate that long-term exposure to the environment with severe air pollution is associated with CFR of COVID-19. Air pollutants such as PM2.5 may have potential ability to predict the CFR of COVID-19.


Background
The novel coronavirus (COVID-19) was rst reported in Wuhan, China in December 2019 [1,2], then spread rapidly worldwide. It was declared as pandemic by the World Health Organization (WHO) in March 2020 [3]. Recently, China is keeping the virus within a controllable range and social order gradually resuming normal. However, other countries and regions around the world are still being threatened by this virus, and the number of con rmed cases and deaths is increasing tremendously. According to the data from WHO, the mortality rate of this epidemic varies in different countries. Now, COVID-19 is a huge obstacle to the world economy and people's normal life. There is no doubt that exploring the relevant factors that affect the mortality rate of the COVID-19 is urgent and vital. It will improve the con dence and means of doctors to treat the virus and reduce the public's terror.
Previous studies found that the morbidity and mortality of COVID-19 was related to individual characteristics, such as gender, age, hypertension, diabetes, and obesity [4][5][6][7][8]. In addition, as one of the triggers of respiratory and systemic in ammatory responses, air pollution should also be noted. The primary visceral target organ of COVID-19 virus and air pollution is also the lung. Previous studies have demonstrated that air pollution increased the incidence of airway in ammation and lung diseases [9,10]. Recent research proved that short-term exposure to air pollution will increase the risk of COVID-19 infection [11][12][13]. The impact of air pollution has a chronic cumulative process on people. Until now, there are only a few relevant research on the relationship between the long-term air pollution and the outcome of patients with COVID-19, and no consistent conclusions [14,15]. We speculate that long-term exposure to air pollution will affect the case fatality rate (CFR) of patients with COVID-19. Therefore, we attempted to explore this relationship in 14 cities in China.

Data collection
Inclusion criteria: (I) The city is a capital city or a municipality directly under the Central Government; (II) Until April 28th, 2020, the number of con rmed cases of COVID-19 is not less than 100 in the city; (III) The air quality data of the city is available on the China Air Quality Online Monitoring and Analysis Platform (https://www.aqistudy.cn).
The number of con rmed cases, deaths and discharges during the epidemic period in the city were recorded from the o cial websites of national or urban Health Commissions (http://www.nhc.gov.cn). In addition, the CFR of COVID-19 were calculated during the current epidemic in the city (Case Fatality Rate = number of deaths / (number of deaths + number of discharges)).
Data of Air Quality Index (AQI), PM2.5, PM10, SO 2 , NO 2 and O 3 in the same period of the city were collected, and the average values in the past 1 year, 3 years and 5 years were calculated respectively. Statistical analysis SPSS 25.0 was used to perform all statistical analyses. Shapiro-Wilk test was used to check whether indicator conforms to the normal distribution. A simple scatter plot was used to illustrate the relationship between air quality indicators of 1 year, 3 years, and 5 years and the CFR.
The Pearson correlation test was performed for normally distributed variables, otherwise the Spearman test was used. According to the CFR of China (except Hubei Province), the cohort was divided into a low CFR group (CFR < 0.86%; n = 6) and a high CFR group (CFR > 0.86%; n = 8). All continuous variables were presented as mean ± standard deviation and compared with the independent-samples t test for normally distributed variables. In addition, ROC curve was further determined. An alpha level of 0.05 was considered statistically signi cant.

Results
In this study, there were 14 cities included in China (Fig. 1
Comparison of air quality between high CFR group and low CFR group As of April 28th, 2020, 13962 cases were cured in China (except Hubei Province), with a total of 121 deaths. The national CFR (except Hubei Province) was 0.86%. According to this index, our cohort was divided into a low CFR group (CFR ≤ 0.86%; n = 6) and a high CFR group (CFR > 0.86%; n = 8). The list and geographical distribution of each group was shown in Fig. 1.
However, there was no signi cant difference in the PM10, SO 2 , NO 2 , and O 3 between high CFR group and low CFR group (P > 0.05).  (Table 4 and Fig. 4).

Discussion
In this multi-city study,we try to estimate the association between long-term exposures to key ambient air pollutants and COVID-19 CFR, and calculate their predictive values. We observed that AQI, PM2.5 and PM10, but NO 2 and O 3 , have showed signi cant positive correlation with CFR.
Meanwhile, compared with low CFR group, AQI and PM2.5 were signi cantly higher in high CFR group. Furthermore, after analysis the predictive value of air pollution indicator to CFR by ROC curve, Long-term PM2.5 may be the potential one. These ndings help for understanding the different severe COVID-19 outcomes between various region. As for this pandemic, stronger healthcare and public attention should be focus on these vulnerable populations, who has suffered from long-term severe air pollution.
Studies have showed that environment play an important role on the transmission and outcomes of some major pandemic and epidemic, namely Severe Acute Respiratory Syndrome (SARS) [16], H1N1 in uenza [17]. The role of the atmospheric environment in COVID-19 was not completely clear [18]. Frontera [19] mentioned that the outbreak of COVID-19 may have an potential inevitable association with climatic conditions and air pollutants. After that, Large number of scholars nd that air pollutants are substantially associated with an increased risk of COVID-19 [11,13,[20][21][22]. For example, according to Zhu's study of 120 city in China, COVID-19 infection is closely related to the short-term exposure to PM2.5, PM10, CO, NO 2 and O 3 [11]. However, the evidences of COVID-19 outcomes and Long-term air pollutants is insu cient. So, the results of our research have important signi cance. We observed signi cant positive associations between long-term PM2.5 levels and COVID-19 CFR. It's well known that PM2.5 affect the airways and the cardiovascular system, it can penetrate deep into lung tissue, causing immune and in ammation response [23][24][25]. Long-time exposure has been correlated with increased risks of chronic diseases, including chronic obstructive pulmonary disease (COPD) that can develop into lung cancer [26,27]. In addition, short-term exposure to PM2.5 is robust associate with increased mortality from various cardiopulmonary diseases, from a nationwide analysis in 272 chinese cities [28]. Considering that COVID-19 and respiratory system are closely related, there is ample reason to believe that PM2.5, on the basis of eliciting immune or in ammatory response, will affect the CFR of COVID-19.
It's inevitably to gure out that NO 2 , which was null associations with severe outcomes of COVID-19 in our research, should also be taken seriously. Ogen [29] found that long-term exposure to NO 2 may be an important cause of death with COVID-19 in Italy and Spain. At the same time, Liang and Shi [14]conducted a cross-sectional nationwide study to estimate the association between long-term county-level exposures to NO 2 , PM2.5 and O 3 and county-level COVID-19 case-fatality and mortality rates in the US, and found that Long-term exposure to NO 2 may enhance susceptibility to severe COVID-19 outcomes. However, all of those studies come from Europe or the USA. As the rst nationwide study in China to estimate the relationship between long-term exposure to air pollutants and COVID-19 CFR, the regional and ethnic differences should not be ignored.
In our study, in the 14 cities, most air pollutants such as NO 2 , PM2.5 and PM10 have gradually decreased over the past ve years, as the Fig. 3 shown. This is attributed to the increased awareness of environmental protection. However, considering that the impact of air pollutants on people's lives is through eliciting immune or/and in ammatory responses. Long-term air pollutants have a cumulative effect(which is inevitable), and most of the current studies on air pollution and COVID-19 focus on short-term air pollution, which is inappropriate and incomplete for this aspect of research.
It could not be ignored that there were still some de ciencies in this study. Firstly, factors and ethnic characteristics that may affect the case fatality rate were not included such as population density, medical level, government policies, Secondly, although our study provided a reference for studying the relationship between long-term air pollution and CFR, our study was limited to China and further research is needed. Thirdly, as this was a retrospective cohort study with a relatively smaller sample size, the sampling error was hard to avoid.

Conclusion
Long-term exposure to the environment with severe air pollution is associated with CFR of COVID-19. Air quality indicators such as PM2.5 may have potential ability to predict the CFR of COVID-19.  " (1-year)" is the abbreviation of "Average in the past year"; " (3-year)" is the abbreviation of "Average in the past 3 years"; " (5-year)" is the abbreviation of "Average in the past 5 years"   * indicates that this data is statistically signi cant; "(1-year)" is the abbreviation of "Average in the past year"; "(3-year)" is the abbreviation of "Average in the past 3 years"; "(5-year)" is the abbreviation of "Average in the past 5 years"; * indicates that this data is statistically signi cant; "(1-year)" is the abbreviation of "Average in the past year"; "(3-year)" is the abbreviation of "Average in the past 3 years"; "(5-year)" is the abbreviation of "Average in the past 5 years"; Figure 1 Geographic patterns of COVID-19 CFR in the 14 cities in China till April 28th, 2020. CFR: Case Fatality Rate Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.