The Effects of Air Pollutants, Particulate Matter 10 (PM10), Sulphur Dioxide (SO2) and Nitrogen Dioxide (NO2) on COVID-19 cases in Indonesia

ABSTRACT

This study aims to analyse the effectsof air pollutants on the number of COVID-19 cases in Indonesia. Three pollutants, i.e. Particulate matter 10 (PM10), Sulphur dioxide (SO2) and Nitrogen dioxide (NO2), were analysed. The studycovers a period of 1 March 2020 to31 December 2020 involving data from the cities ofJakarta, Bandung, Yogyakarta, Semarang and Surabayain Indonesia.

This studyused the OrdinaryLeast Square (OLS) method with the endurance test Robust Standard Errors. The regression results showed that PM10, SO2and NO2 are statistically significant positive regressors of the number of COVID19 cases. Every1 μg/m3 increase in PM10, SO2 and NO2concentrationsis shown to causean additional2.65, 7.96 and 21.01 cases of COVID-19, respectively. The implementation of Large-Scale Social Restrictions (PSBB) has a statistically significant impact incurbingCOVID-19 transmission; reducing447.4 cases of COVID-19.

Keywords: COVID-19, PM10, NO2, SO2, regression

INTRODUCTION

Coronavirus disease 2019 (COVID-19) or Several Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) first appeared in Wuhan City, China, in December 2019 (Jain et al.,2000). COVID-19 spread swiftly around the world, including in Indonesia. Data from the Ministry of HealthIndonesia showed that until 24 November 2021, the total number of COVID-19 cases worldwide reached 258,164,425 caseswith 5,166,192 deaths. In Indonesiaas of 24 November 2021, a total of 4,254,443 people were confirmed positive for COVID-19 with 143,766 deaths and a Case Fatality Rate (CFR)of 3.4%. As many as 4,102,700 people managed to recover from COVID-19.

The Indonesian government applied some measures to curbCOVID-19spread, one of them is the implementation of Large-Scale Social Restrictions, or Pembatasan Sosial Berskala Besar (PSBB) in Indonesian language.This research aimsto identify the effectsof air pollutants on the rise of COVID-19 cases is Indonesia during the implementation of PSBB. Studies analysing how air pollutants affect the number of COVID-19 cases and deaths, as well as that ofother airborne diseases such as Sever Acute Respiatory Syndrome (SARS) are relatively voluminous. These studiesusedcities in China, Italy, the United States of America and Europe Unionas the cases. See for example Contini and Costabile (2020), Cui et al(2003), Gautam (2020), Setti et al (2020), ..Such a study for Indonesian cities, however, is still relatively scarce, motivating the author to conduct such a study for five large Indonesian cities, i.e. Jakarta, Bandung, Yogyakarta, Semarang and Surabaya. The researcher hopes that resultsof this research will give some contributionsto decision-makers in designing regulations, especially the ones that are related to environmental factors that affect COVID-19 cases and death due to COVID-19.

On 2 January2020, 41 people treated in hospitals by laboratoriesin Wuhan were confirmed positive for COVID-19. As many as 30 out of 41 people (73 percent) were male, of which 13 people had congenital disease: 8 people had diabetes (20%), 6 people had hypertense(15%) and 6 people had cardiovascular (15%). The average age of the aforementioned people was 41 years old. Of that numbers, 13 of 41 people were treated in the ICU and 6 of them were deceased. Of the 13 people who had to be treated in the ICU had high plasma levels (Huang et al.,2020).

The death rate due to COVID-19 increased. On January 24th2020, laboratoriesin Wuhan stated that there were 835 patients infected by COVID-19 with 25 cases of them declared fatal. Clinical symptoms in those patients were similar to patients with SARS-CoV. The patients suffered from fever with cough, shortness of breath, fatigue, headache and diarrhoea. All mentioned patients also had pneumonia with abnormal chest CT including acute respiratory distress. From the tracing result, it was known that there was a family of the patients who consumed seafood from Huanan Seafood Market prior they were confirmed positive for COVID-19 (Huang et al.,2020). Based on the progress of the tracing, 27 people were exposed to COVID-19 after they went to Huanan Seafood Market. Scientists then concluded that Huanan Seafood Market in Wuhan, China was the first place that COVID-19 appeared. Bat meat, one of the products sold in the Huanan market, is taken as the medium of COVID-19’s transmission. The transmission from one person to others happened because of physical contact and droplets (Zhou et al.,2021).

From Wuhan in China, COVID-19 spread swiftly around the globe. The World Health Organisation (WHO) declared the COVID-19 epidemic a Public Health Emergency of International Concern (PHEIC) on 31 January 2020 (Zhou et al.,2021). On 11 March 2020, WHOdeclared the COVID-19 a pandemic due to its spreading to several countries, like Nepal, India and some countries in South Asia.

Indonesia was officially included as countries infected by COVID-19 on2March 2020. The first case was found in two people from Depok, West Java. From tracing results, it was suspected that they were infected by COVID-19 after getting physical contact with a Japanese citizen that visited Jakarta. The Japanese citizen was known to be infected by COVID-19 in Malaysia on his way home from Indonesia to Japan.

Some measures were takenby the Government of Indonesia to prevent the spread of COVID-19, including the PSBB andand vaccination. Since August 2021, the trend of people infected by COVID-19 started to decrease. However, it does not mean that the spread has stopped. Countries worldwide, including Indonesia, still need to face Omicron, a new variant of SARS-CoV-2 or COVID-19.

The escalation of infection rate and death rate due to COVID-19 since the pandemic first appeared in Wuhan, China, in December 2019 and spread around the world has intrigued some researchers (Jain et al.,2000). Researchers tried to analyse factors related to the spread of COVID-19. Some researchers found that COVID-19 transmits with direct physical contact and droplets from breathing.In some research, some researchers used air pollutants as the focus of their research. The consideration of using air pollutants is that the increase in air pollution is a threat to human health, especially related to respiratory disease (He Li et al.,2020). Several previous researchhad proved that there is a correlation between air pollution in the form of particulate matter in short term exposure to ambient air towards the increased risk of respiratory disease. The increased risk happens because particulate matter in the air consists of toxic substances that could get in from the inhalation process and later will circulate in human’s bloodstream and targeted organs, which will affect the exposed person (Firmansyah et al., 2020).

Previous studies reportedthat pollutants like Nitrogen dioxide (NO2), Sulphur dioxide (SO2), Carbon monoxide (CO), Particulate matter 2.5 (PM2.5) and Particulate matter 10 (PM10) were the cause of cardiopulmonary disease. Nitrogen dioxide (NO2) leads to upper respiratory tract infections (URTI). Sulphur dioxide (SO2) and Carbon monoxide (CO) could lead to an increase in the risk of stroke, asthma and also lung cancer (Banerji, S.andMitra,D.,2021).

Pollutants of Particulate matter 2.5 (PM2.5) and Particulate matter (PM10) with a diameter less than 0,5 mm can levitate in the air at a far distance. If the particles in the air are inhaled by humans, they can be distributed to the lungs and cause inflammation, oxidative stress and lung cancer (Valavanidis et al.,2008; Anderson et al.,2012; Kim et al.,2015). Furthermore, several researchalso found that there is a correlation between air pollution and infectious disease transmission (He Li et al.,2020), in example: worst air quality has been proven could increase the death due to SARS and increase thenumber of influenzas (Cui et al., 2003; Landguth et al., 2020).

The data showed that the air quality had significantly and positively affected the increase of daily cases of COVID-19 in Wuhan, China. Pollutants Particulate matter 2.5 (PM2.5) and Particulate matter 10 (PM10) became transmission media of SARS-CoV-2 andpotentially increased the spread of COVID-19 (He Li et al., 2020).

Particulate matter (PM) has a toxic effect that could enter human lungs and affect the physiological condition of human lungs alsoincrease the risk of mortality and morbidity of COVID-19. Some researchesfound that there is a correlation between PM and COVID-19. PM2.5, which is smaller than PM10, has a higher ability to become a weighting factor in COVID-19 (Firmansyah et al.2020). There are two ways of COVID-19 transmission through PM. First, PM2.5 blocks the human respiratory process and second, PM can form condensation nuclei for virus attachment (Lee et al., 2014). PM2.5 is relatively smaller in size so that it can penetrate alveoli and damage the respiratory tract. PM2.5 became the most dominantfactor that could transmit SARS-Cov-2 or COVID-19 (He Li et al., 2020).

Research discussing the effectsof pollutants on the increase of COVID-19 cases has been done in cities in several countries, like China, Italy and the USA. However, the aforementioned research has notbeen done in Indonesia. Through this research, it is hoped that the effect of pollutants on the increase in COVID-19 cases will be known, especially during the implementation of PSBB.

This research aimed to analyse the influence of air pollution on the increase of COVID-19 cases in Indonesia during the implementation of Large-Scale social restrictions (PSBB). The health of people that confirmed positive COVID-19 can get worse when exposed to air pollutants that to date always been known as one of thereasons forrespiratory disease (Firmansyah et al., 2020). However, currently, only a few people know about the influence of environmental factors on the increase of COVID-19 cases, even though there is some amount of evidence that the increase in air pollution rate in statisticscould have increased the positive confirmed COVID-19 case and death due to COVID-19 (Persico et al.,2021).

METHODS

Data for this study were obtained from Kementerian Kesehatan (n.d.) for the period of 1 March 2020 to 31 December 2020 in five big cities in Indonesia, i.e. Jakarta, Bandung, Yogyakarta, Semarang and Surabaya. During this period, several phases of Large-Scale Social Restrictions (PSBB) was implemented. Variables used in this research are shown in Table 1.

Table 1.Research variables

Descriptive statistical variables in this research are shown in Table 2. The first panel shows COVID-19 as a dependent variable (Y) consistingof 1,408 cases with an average of 783.08daily cases. Variable X in this research are pollutant Particulate matter 10 (PM10) with a total observation is 1,513 and average daily concentrate is 32,38 μg/m3, pollutant Sulphur dioxide (SO2) with a total observation is 1,452 and average daily concentrate is 24.23μg/m3, and pollutant Nitrogen dioxide (NO2) with total observation is 1,149 and average of daily concentrate is 15.95 μg/m3 as shown in Table 2.

Table 2.Descriptive statistics

The objective of this research is to test the influence of air pollutants: Particulate matter 10 (PM10), Nitrogen dioxide (NO2) and Sulphur dioxide (SO2) on the increase of COVID-19 casesin Indonesia using the Ordinary Least Square (OLS) method. The analysis was started by composing a descriptive statistic from the data that will be used in the double linear regression model. The goal of this descriptive statistic is to decide the base feature from datasets of the observed data. There are three methods to estimate the panel data (Baltagi, 2005; Wooldridge, 2016). The three methods are Common Effect (CE), Fixed Effect (FE) and Random Effect (RE). Of the three models, the researcher picked one that can be used as the model for estimating panel data parameters.

Later on, the authordid three tests to choose whether to use Common Effect, Fixed Effect or Random Effect. The tests are F Statistic Test (Chow Test) to choose between the Common Effect or Fixed Effect method. Hausman Test to choose between Fixed Effect or Random Effect and Lagrange Multiplier (LM) Test to choose between Common Effect or Random Effect.

The regression model used in this research has autocorrelation, heteroscedasticity and normality. In econometrics, this terminology is known as Heteroscedasticity Autocorrelation Spatial Correlation (HACSC) (Vogelsang, 2012). There are several methods to solve this problem, Beck & Katz (1995) suggested using panel-corrected standard errors (PCSE) (Hoechle, 2007). Monte Carlo analysis mentioned that panel-corrected standard errors is suitable for the regression model.

Robust Standard Errors can also be used in a regression model. Robust regression is essential to solve the problem in Ordinary Least Squares (OLS), like autocorrelation, heteroscedasticity and normality (Alma,2011). Robust Standard Errors is usuallyapplied in cross-section regression, especially in larger data. Even nowadays, it is not rare that researchers implement Robust Standard Errors so it will be easier to get a result that is more resistant to the problems occurring in OLS regression. In this research, the researcher used Robust Standard Errors to overcome the autocorrelation, heteroskedasticity and normality problems.

Based on the data that the researcher received and the objective of this research, that is to analyse the influence of air pollutants on the increase of COVID-19 cases, then the double econometric linear regression model that the researcher used was:

Yit = α + β1PM10it + β2SO2it + β3NO2it + DummyPSBBit + humidityit + ɛit

Y is confirmed positive COVID-19 cases, α is constant, β is coefficient, ɛis an error term, i is the observed city and t is the period of observation. PM10, SO2 and NO2 are the pollutants in μg/m3 which become independent variables or X, PSBB is Large-Scale Social Restriction during COVID-19 pandemic.

RESULTS AND DISCUSSION

This research’s objective is to know the effectsof air pollutants, which are Pollutant Matter 10 (PM10), Sulphur dioxide (SO2) and Nitrogen dioxide (NO2), on the number ofCOVID-19 cases in Jakarta, Bandung, Yogyakarta, Semarang and Surabaya. Data were obtained from Ministry of Health’s official records, covering the period of 1March 2020 to 31December 2020(Kemenkes, n.d.). The regression result from Ordinary Least Squares (OLS) is shown in Table 3.

Table 3.The effects of PM10, SO2 and NO2 on COVID-19 cases

Notes:** = statistically significance at α= 5%

*** = statistically significance at α= 2.5%

As shown in Table 3, PM10 air-concentration is a statistically significant positive regressor for the number ofCOVID-19 cases;every 1 μg/m3 increase in PM10 concentration could lead to2.65 additional COVID-19 cases. SO2 is also shown to be a statistically significantpositive regressor forthe number of COVID-19 cases; every 1 μg/m3 increase of SO2 concentration add another 7.96 COVID-19 cases. So is NO2, where an increase of NO2 concentration by 1 μg/m3increasesthe number of COVID-19 casesby 21.01.

On the contrary, PSBB is shown to be statistically significant negative regressor of the number of COVID-19 cases, in which the implementation of PSBB decreases the number of cases by 447.4.

Robustness Check

The next step was a robustness check or endurance test. In this step, the data was analysed with a similar model as pollutant datasets, which are PM10, SO2 and NO2. The researcher conducted this test to see how far the consistency of the influence of PM10, SO2and NO2 is towards the increase of COVID-19 cases from the first day until the seventh day after the observation. The result of the robustness check is shown in Table 4.

The results showed that there is a statistical consistency in the influence of PM10, SO2 and NO2 on the number of COVID-19 cases. There is also a variation of significant rate between the three pollutants. The effects of PM10on the first day is shown to be statistically significant, with the statistical significance becoming stronger on the second tothe seventh day. Pollutants SO2and NO2 are shown to be significant regressors of the number of COVID-19 cases at α= 2.5%.The PSBB is statistically significant in decreasing COVID-19 cases to 479.7 cases in the first day.

Table 4.Robustness check

Notes:** = statistically significance at α= 5%

*** = statistically significance at α= 2.5%

CONCLUSIONS

The objective of this research is to analyse the effectsof Particulate matter 10 (PM10), Sulphur dioxide (SO2) and Nitrogen dioxide (NO2) on the numberof COVID-19 cases.The results showedthat PM10, SO2and NO2 had a statistically significant effectsonthe numberof COVID-19 cases. Theseresults are consistent with thoseof Bashir et al., 2020 who found a strong correlation between air pollutants, PM10, PM2.5, SO2, NO2 and CO and the COVID-19 pandemic in California, USA.

This study’s finding that NO2 statistically significantly affectthe number of COVID-19 casesoffers an additional support to Ogen (2020)’s finding that3,487 (78%) out of 4,443 COVID-19 deaths in Italy, Spain, France and Germany occurred inareas with a high rate of NO2 concentration, such as in Northern Italy and Spain’s city centres.

This study, however, does not use pollutant sources, e.g. household and industrial energy consumptionand motor vehicle emission, as explanatory variables. A future analysis of how pollutant sources affect the transmission of airborne diseases would be very useful for both academic and policy-making purposes.

REFERENCES

Alma, O.G. (2011), Comparison of Robust Regression Methods in Linear Regression. International Journal Contemporary Mathematical Sciences 6(9): 409—421.

Anderson,J.O., Thundiyil JG, and Stolbach,A.(2012), Clearing the Air: A Review of the Effects of Particulate Matter Air Pollution on Human Health. Journal of Medical Toxicology8(2): 166—175. DOI: https://doi.org/10.1007/s13181-011-0203-1

Badi H. Baltagi (2005), Econometric Analysis of Panel Data. Third edition. Chichester: John Wiley & Sons.

Banerji, S., and Mitra, D. (2021), Assessment of Air Quality in Kolkata Before and After COVID-19 Lockdown. Geocarto International: 1—24. DOI:10.1080/10106049.2021.1936209

Beck, N., and Katz, J.N. (1995), What to do (and not to do) with Time-Series Cross-Section Data. The American Political Science Review89(3): 634—647. DOI: https://doi.org/10.2307/2082979

Contini, D., and Costabile, F. (2020), Does Air Pollution Influence COVID-19 Outbreaks?.Atmosphere Journal11(4): 377. DOI: https://doi.org/10.3390/atmos11040377

Cui, Y., Zhang, Z.F., Froines, J., Zhao, J., Wang, H., Yu, S.Z., and Detels, R. (2003), Air Pollution and Case Fatality of SARS in the People's Republic of China: An Ecologic Study. Environmental Health2(1): 15. DOI: https://doi.org/10.1186/1476-069X-2-15

Firmansyah, H., Fadlillah, A.N., and Pawitra, A.S.(2020), Particulate Matter as Driven Factor Covid-19 Transmission at Outdoor: Review. Journal of Environmental Health12(3): 225—234. DOI: http://dx.doi.org/10.20473/jkl.v12i3.2020.225-234

Gautam,S.(2020), The Influence of COVID–19 on Air Quality in India: ABoon or Inutile. Bulletin of Environmental Contamination and Toxicology 104(6): 724—726. DOI: 10.1007/s00128-020-02877-y

He Li, Xiao-Long Xu, Da-Wei Dai, Zhen-Yu Huang, Zhuang Ma, and Yan-Jun Guan (2020), Air Pollution and Temperature are Associated with Increased COVID-19 Incidence: A Time Series Study. International Journal of Infectious Disease 97: 278—282. DOI: 10.1016/j.ijid.2020.05.076

Hoechle, D. (2007), Robust Standard Errors for Panel Regressions with Cross-Sectional Dependence. The StataJournal7(3): 281—312. DOI: https://doi.org/10.1177/1536867X0700700301

Huang, C., Wang, Y., Li, X., Ren, L., Zaho, J., Hu, Y., Zhang, L., Fan, G, Xu, J., Gu, X., Cheng, Z., Yu, T., Xia, J., Wei, Y., Wu, W., Xie, X., Yin, W., Li, H., Liu, M., Xiao, Y., Gao, Z., Jin, Q., Wang, J., Cao, B. (2020), Clinical Features of Patients Infected with 2019 Novel Coronavirus in Wuhan, China. The Lancet 395. DOI: 10.1016/S0140-6736(20)30183-5

Jain, N., Choudhury A., Sharma, J., Kumar, V., De D., Tiwari, R.A. (2020), A Review of Novel Coronavirus Infection (Coronavirus Disease-19). Global Journal of Transfusion Medicine 5:22—26. DOI: 10.4103/GJTM.GJTM_24_20

Kementerian Kesehatan (n.d.). COVID-19 Indonesia. Available at: https://covid19.kemkes.go.id/. [Accessed from 18 August 2021]

Kim, B.M., Park, J.S, Kim, S.W., Kim, H. (2015), Source Apportionment of PM10 Mass and Particulate Carbon in the Kathmandu Valley, Nepal. Atmospheric Environment 123: 190—199. DOI:10.1016/j.atmosenv.2015.10.082

Landguth,E., Holden,Z., Graham,J., Stark,B., Mokhtari,B.E., Kaleczyc,E., Anderson, S., Urbanski, S., Jolly, M., Semmens, E.O., Warren, D.A., Swanson, A., Stone, E., Noonan, C.(2020), The Delayed Effect of Wildfire Season Particulate Matter on Subsequent Influenza Season in a Mountain West Region of the USA. Environmental International 139. DOI: https://doi.org/10.1016/j.envint.2020.105668

Lee B.J., Kim B., and Lee K. (2014), Air Pollution Exposure and Cardiovascular Disease. Toxicological Research 30(2): 71—75. DOI: 10.5487/tr.2014.30.2.071

Persico, C.L.and Johnson, K.R. (2021), The Effects of Increased Pollution on COVID-19 Cases and Deaths. Journal of Environmental Economics and Management 107. DOI: 10.1016/j.jeem.2021.102431

Setti, L., Passarini, F., Gennaro, G.D, Barbieri, P., Perrone, M.G., Piazzalunga, A., Borelli, M., Palmisani, J., Gilio, A.D, Piscitelli, P., Miani, A. (2020), Potential Role of Particulate Matter in the Spreading of COVID-19 in Northern Italy: First Evidence-based Research Hypotheses. BMJ Open10. DOI:https://doi.org/10.1101/2020.04.11.20061713

Setti, L., Passarini, F., Gianluigi, G.D., Gillo A.D., Palmissani, J., Buono, P., Fornari, G., Perrone, M.G., Piazzalunga, A., Barbieri, P., Rizzo, E., Miani, A. (2020), Evaluation of the Potential Relationship between Particulate Matter (PM) Pollution and COVID-19 Infection Spread in Italy. Accessible at: https://www.ptpz.pl/wp-content/uploads/2020/04/COVID_19_position-paper_ENG1.pdf

Vogelsang, T.J.(2012), Heteroskedasticity, Autocorrelation, and Spatial Correlation Robust Inference in Linear Panel Models with Fixed-Effects. Journal of Econometrics166(2): 303—319. DOI:10.1016/j.jeconom.2011.10.001

Valavanidis,A., Fiotakis,K., and Vlachogianni,T.(2008), Airborne Particulate Matter and Human Health: Toxicological Assessment and Importance of Size and Composition of Particles for Oxidative Damage and Carcinogenic Mechanisms. Journal Environmental Science and Health 26(4): 339—362. DOI:10.1080/10590500802494538

Wooldridge, J.M. (2016), Introductory Econometrics: A Modern Approach. 6th edition. Boston: Cengage Learing.

Zhou, L., Ayeh, S.K., Chidambaram, V., and Karakousis, P.C. (2021), Modes of Transmission of SARS-CoV-2 and Evidence for Preventive Behavioral Interventions. BMC Infectious Diseases 21(496). DOI: https://doi.org/10.1186/s12879-021-06222-4