key: cord-0968360-mz7b6ey3
authors: Lee, Sang Chul; Son, Kang Ju; Kim, Dong Wook; Han, Chang Hoon; Choi, Yoon Jung; Kim, Seong Woo; Park, Seon Cheol
title: Smoking and the risk of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection
date: 2021-04-23
journal: Nicotine Tob Res
DOI: 10.1093/ntr/ntab079
sha: c97f9a6e2d7285ac00d56b1beef59461b7123e96
doc_id: 968360
cord_uid: mz7b6ey3

INTRODUCTION: It is unclear whether smokers are more vulnerable to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. This study aimed to evaluate the association between smoking and the risk of SARS-CoV-2 infection. METHODS: A matched case-control study was conducted using a large nationwide database. The case group included patients with SARS-CoV-2 infection confirmed by the Korea Centers for Disease Control and Prevention, and the control group was randomly sampled from the general Korean population in the National Health Insurance Service database by matching sex, age, and region of residence. Conditional logistic regression models were used to investigate whether the risk of infection with SARS-CoV-2 was affected by smoking status. RESULTS: A total of 4,167 patients with SARS-CoV-2 infection and 20,937 matched controls were enrolled. The proportion of ex-smokers and current smokers was 26.6% of the total participants. In multivariate analysis, smoking was not associated with an increased risk of SARS-CoV-2 infection (odds ratio [OR] = 0.56, confidence interval [CI] = 0.50–0.62). When ex-smokers and current smokers were analysed separately, similar results were obtained (current smoker OR = 0.33, CI = 0.28–0.38; ex-smoker OR = 0.81, CI = 0.72–0.91). CONCLUSIONS: This study showed that smoking may not be associated with an increased risk of SARS-CoV-2 infection. Smoking tends to lower the risk of SARS-CoV-2 infection; however, these findings should be interpreted with caution. IMPLICATIONS: It is unclear whether smokers are more vulnerable to coronavirus disease 2019. In this large nationwide study in South Korea, smoking tended to lower the risk of infection with severe acute respiratory syndrome coronavirus 2. However, these findings should be interpreted with caution, and further confirmatory studies are required.

Coronavirus disease is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The COVID-19 outbreak started in Wuhan, Hubei Province, China, in December 2019 and has been spreading globally since then. As of March 9, 2020, a total of 116,166,652 COVID-19 cases have been reported worldwide, including 2,582,528 deaths. 1 Cigarette smoking increases the risk of important bacterial and viral infections through structural changes in the respiratory tract and a decrease in the immune response. 2 During the previous SARS outbreak (2002) (2003) (2004) , there was insufficient data on smoking and infection risk. One study reported that smoking did not increase the infection risk; however, no further studies have confirmed this. 3 The invasion and replication of SARS-CoV-2 mainly occur in the respiratory tract. 4 Smoking weakens the mucociliary clearance and immune system of the airway, and makes SARS-CoV-2 easily penetrate the host cell. 5 Moreover, active cigarette smoking upregulates angiotensin-converting enzyme (ACE) 2, which is the host receptor for SARS-CoV-2 in the lower airway; thus, smoking may increase the risk of SARS-CoV-2 infection. 6, 7 However, it is not well known whether smokers are more likely to get infected by SARS-CoV-2 or are more vulnerable to COVID-19. This study aimed to evaluate the association between smoking and the risk of SARS-CoV-2 infection.

A c c e p t e d M a n u s c r i p t 6 

We used the Korean National Health Insurance Data and COVID-19 registry data from the Korea Centers for Disease Control and Prevention (KCDC). Korea has a unique insurance system provided by the National Health Insurance Service (NHIS). Nearly all individuals living in Korea are obliged to join the NHIS, and a large amount of medical information has been accumulated in the NHIS database. Recently, the NHIS provided the nationwide data on COVID-19 for research collaboration. The NHIS dataset for research on COVID-19 was based on insurance claims and comprised patients with SARS-CoV-2 infection, matched with the general population. This dataset includes various health-related variables, including sociodemographic information, diagnosis, medication, national health screening information (including smoking survey), and survival status. This study was approved by the Institutional Review Board of the NHIS Ilsan Hospital, Korea, and adhered to the tenets of the Declaration of Helsinki (NHIMC 2020-04-026). Owing to the retrospective nature of the study and the use of de-identified patient data from the database, the requirement for written informed consent was waived.

The COVID-19 registry data of the KCDC includes all COVID-19 cases in Korea to date.

The first COVID-19 case in Korea was confirmed on January 20, 2020. The case group included patients with SARS-CoV-2 infection confirmed by the KCDC until April 16, 2020.

All diagnoses in the registry data of the KCDC were confirmed by reverse transcription polymerase chain reaction testing for SARS-CoV-2. Patients under 40 years of age were A c c e p t e d M a n u s c r i p t 7 excluded because smoking information was not available at this age. The control group was randomly sampled from the general Korean population in the NHIS database by matching sex, age, and the region of residence, and the number of the control group was set to be five times that of the case group. Because all COVID-19 cases in Korea were included in the KCDC registry, it was considered that the control group did not have COVID-19.

In both case and control groups, information on sex, age, region of residence, and household The NHIS dataset includes information on national health screening, which is mandatory for most Korean citizens every two years. Personal smoking information was obtained through a smoking survey included in the national health screening program. The most recent health screening data prior to the time of enrollment were used to obtain smoking information.

Smoking information was obtained through the following questionnaire: "Have you ever smoked more than five packs of cigarettes in your life?" The answers were divided into three categories: no, quit now, or currently smoking. The risk of infection with SARS-CoV-2 was analysed according to smoking status, which was categorized as never-smoker, ever-smoker, and current smoker. 

Descriptive statistics were performed for all variables. Differences between the two groups were assessed using the chi-square test. Conditional logistic regression analyses that were conditioned on sex, age, and region of residence were performed to determine whether smoking was associated with the risk of infection with SARS-CoV-2. Two models were used to analyze the risk of infection: the first model (Model 1) compared two groups (neverversus ever-or current smokers); the other model (Model 2) compared three groups (never versus ever versus current smokers). Other variables (household income and comorbidities), except smoking, were the same for Models 1 and 2. Statistical significance was set at P < 0.05. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).

A total of 4,167 patients with SARS-CoV-2 infection were included ( Table 1 ). The matched control group consisted of 20,937 participants. Of the total subjects, 8,814 (35.1%) were men.

The age distribution was the highest in the 50s (34.0%), and those in their 60s and 40s were 24.7% and 23.3%, respectively. The incidence of SARS-CoV-2 infection was high in Daegu and Gyeongbuk, which are located in the southeastern region of South Korea. In the univariate analysis, lower income levels were associated with the risk of infection with SARS-CoV-2, especially in the Medicaid group. Among the five categories of comorbidities, A c c e p t e d M a n u s c r i p t 9 respiratory diseases were more frequent in patients with SARS-CoV-2 infection, while hypertension was less frequent. Heart disease, diabetes mellitus, and cancer did not differ between the case and control groups.

The proportion of smokers (ex and current) was 26.6% of the total subjects and 21.6% and 27.6% in the case and control groups, respectively. In the conditional logistic regression analysis, smoking was not associated with an increased risk of SARS-CoV-2 infection (Table 3) .

This large population-based study found that smoking was associated with a reduced risk of SARS-CoV-2 infection. These results were the same for both current and ex-smokers, even after adjusting for all available confounding variables. However, these findings should be interpreted cautiously.

A c c e p t e d M a n u s c r i p t 10 Early epidemiologic studies of COVID-19 in China have revealed that the proportion of smokers is larger than that of nonsmokers in the general population. 8-10 Subsequently, a cross-sectional study revealed that active smoking was linked with decreased odds of a positive COVID-19 result, 11 and another study from the US and China also found a lower prevalence of smokers in hospitalized patients with COVID-19. [12] [13] [14] This may indicate that the risk of SARS-CoV-2 infection in smokers is not higher than that in non-smokers.

Further studies have reported that smoking was associated with a lower risk of SARS-CoV-2 infection. A cohort study involving aircraft carrier crews found that current smokers had a lower OR of 0.64 (95% CI 0.49-0.84) for COVID-19 occurrence compared to that of nonsmokers. 15 Another study including 38 European countries showed a negative association between smoking prevalence and COVID-19 after controlling for confounding factors. 16 One study using a national internet-based survey in Italy found a dose-dependent decrease in SARS-CoV-2 infection in current smokers. In this study, the OR for SARS-CoV-2 infection however, former smokers did not have a significant association. 18 Although direct comparison is limited due to differences in study methods, these patterns of previous studies are quite similar to those of our study.

These findings have suggested a "smoker"s paradox", and there are possible mechanisms by which smoking has potential protective effects on the severity of COVID-19. 19 The protective effects of smoking include an anti-inflammatory effect of nicotine, 20 a blunted immune response in smokers, 21 and increased nitric oxide in the respiratory tract. 22 Leung et al. analysed the protein expression of ACE-2 in the resected lung tissue specimen of patients with COVID-19. 6 Although ACE-2 gene expression was higher in active smokers, the protein A c c e p t e d M a n u s c r i p t 11 intensity of epithelial ACE-2 was similar to that of never smokers. This may be one reason why the risk of SARS-CoV-2 infection was not higher in healthy smokers. Moreover, some studies reported that the immune system of smokers is more tolerant and less reactive compared to that of never smokers via attenuation of normal defensive function of the immune system. 23, 24 These studies also revealed that exposure to smoke could modulate the adaptive immune responses and reduce systemic levels of several immune or inflammatory markers compared to those in never smokers.

However, this reverse association between smoking and SARS-CoV-2 infection requires careful interpretation. The method of collecting information related to smoking status and patterns of healthcare utilization might be the reasons why smoking groups showed lower positivity for COVID-19. Because the information on smoking history was based on the national health examination questionnaire, it is possible that the examinee did not provide accurate information or the smoking status changed after the health examination. In addition, smokers with symptoms tend to have high medical utilization rates, but symptom-free healthy smokers tend to have low compliance with drug prescriptions or preventive care screening. [25] [26] [27] [28] Therefore, smokers can get tested more because of their symptoms suggestive of COVID- 19. 11,29 For example, both smoking and COVID-19 can cause cough, and smokers with smoking-related cough can be tested even when they do not have COVID-19. 30 In addition, testing for acute infection requires collecting swabs of the mucosal epithelium, which can be disrupted in smokers, and this lead to alter the sensitivity of assays. 11 In Korea, the initial outbreak of COVID-19 had a unique feature. During our study period, a cluster infection in one religious group accounted for half of all patients with COVID-19 in Korea. A high proportion of transmission was due to close contact in the same place, and this feature could overwhelm the individual characteristics of disease susceptibility. Additionally, our study included individuals over 40 years of age, because Korean national health screening A c c e p t e d M a n u s c r i p t 12 was recommended in these age groups. A recent study from the US reported an increased risk of COVID-19 in young smokers. In this study, the risk of COVID-19 was associated with ecigarettes and the dual use of e-cigarettes and cigarettes among young smokers aged 13-24 years. 31 Some previous studies have strongly encouraged smoking cessation to flatten the COVID-19 outbreak and reduce mortality. [32] [33] [34] However, the association between smoking and the severity of COVID-19 remains unclear. A study from Italy demonstrated that active smoking was not associated with COVID-19 severity. 35 In contrast, another study from the UK reported that smokers have a decreased risk of both COVID-19 and ICU admission compared to that of non-smokers. 36 More recent study from the US also found that cumulative exposure to smoking is a risk factor for hospital admission and mortality. 37 Several meta-analyses also reported conflicting results. Lippi et al. reported that active smoking is not associated with the severity of COVID-19 in Chinese patients during the early COVID-19 outbreak. 38 However, an updated meta-analysis using the same Chinese studies found increased severity of COVID-19 in active smokers with a pooled OR of 2.2 (95% CI 1.31-3.67). 39 More recent meta-analyses tend to report that smoking is associated with poor clinical outcomes. 18, 40, 41 When these reports are combined, the risk of poor clinical outcomes in COVID-19 tends to increase in smokers, while the risk of SARS-CoV-2 infection seems to be reduced. However, the decreased risk of infection cannot justify the dangers of smoking itself, such as an increased risk of respiratory infection and progression of COPD or lung cancer. Moreover, there are several studies on high under-diagnosis rate of COPD among smokers in the primary care setting. 42, 43 Thus, possible vulnerability to COVID-19 among smokers who have risk factors for COPD, such as predisposing host factors, > 20 pack-years of smoking, and occupational dust or fume exposure, should be emphasised. 44 Furthermore, the virus can be spread during smoking through crowding in a certain space, talking without wearing a facial A c c e p t e d M a n u s c r i p t 13 mask, bronchial toileting behaviours, such as coughing or spitting phlegm, or hand-to-mouth movement of smokers.

Our study is a large population-based study to evaluate the association between smoking status and the risk of SARS-CoV-2 infection. However, this study has several limitations.

First, as a pandemic, the risk of SARS-CoV-2 infection is highly affected by exposure to the virus. Our study did not capture the exposure status due to the limitations of the data source.

A non-smoker who attended a gathering of a large crowd would have a higher risk of disease than that of a smoker staying at home all the time. Our study could not limit the time window to when social distance ordering was in place to adjust for possible exposure. Second, our study might not reflect the current smoking habits of the subjects at the time of SARS-CoV-2

infection. This study used the patient"s smoking information using a smoking survey included in the national health screening. Therefore, it was difficult to accurately evaluate the smoking status at the time of SARS-CoV-2 infection. Third, our study did not include young COVID- M a n u s c r i p t 22 SARS-CoV-2=severe acute respiratory syndrome coronavirus 2; OR=odds ratio; CI=confidence interval.

World Health Organization. Coronavirus disease (COVID-2019) situation reports 2020

Cigarette smoking and infection

Smoking and Severe Acute Respiratory Syndrome

SARS-CoV-2 Viral Load in Upper Respiratory Specimens of Infected Patients

SARS-CoV-2 COVID-19 susceptibility and lung inflammatory storm by smoking and vaping

ACE-2 expression in the small airway epithelia of smokers and COPD patients: implications for COVID-19

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Clinical characteristics of 140 patients infected with SARS-CoV-2 in Wuhan

Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study

Clinical characteristics of coronavirus disease 2019 in China

Risk factors for SARS-CoV-2 among patients in the Oxford Royal College of General Practitioners Research and Surveillance Centre primary care network: a cross-sectional study

Systematic review of the prevalence of current smoking among hospitalized COVID-19 patients in China: could nicotine be a therapeutic option

Clinical Characteristics of Covid-19 in New York City

Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study

Impact of Tobacco Smoking on the risk of COVID-19.A large scale retrospective cohort study

Smoking Prevalence and COVID-19 in Europe

Current smoking and SARS-CoV-2 infection: findings from the Italian cross-sectional EPICOVID19 internet-based survey

The association of smoking status with SARS-CoV-2 infection, hospitalization and mortality from COVID-19: a living rapid evidence review with Bayesian meta-analyses

Is there a smoker's paradox in COVID-19?

Nicotinic acetylcholine receptor alpha7 subunit is an essential regulator of inflammation

Smoking habit and hospitalization for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-related pneumonia: The unsolved paradox behind the evidence

The role of nitric oxide in cigarette smoking and nicotine addiction

Impacts of cigarette smoking on immune responsiveness: Up and down or upside down?

Cigarette smoking and variations in systemic immune and inflammation markers

The impact of smoking on adherence to treatment for latent tuberculosis infection

The association of smoking status with healthcare utilisation, productivity loss and resulting costs: results from the population-based KORA F4 study

Disparities in health care utilization by smoking status--NHANES 1999-2004

The association of smoking with medical treatment adherence in the workforce of a large employer

Current smoking and COVID-19 risk: results from a population symptom app in over 2.4 million people

Assessing and addressing collider bias in addiction research: the curious case of smoking and COVID-19

Association Between Youth Smoking, Electronic Cigarette Use, and COVID-19

COVID-19 outbreak impact in Spain: A role for tobacco smoking?

The tobacco industry in the time of COVID-19: time to shut it down?

Smoking Cessation as a Public Health Measure to Limit the Coronavirus Disease 2019 Pandemic

Current smoking is not associated with COVID-19

Risk of severe COVID-19 disease with ACE inhibitors and angiotensin receptor blockers: cohort study including 8.3 million people

Association of Smoking and Cumulative Pack-Year Exposure With COVID-19 Outcomes in the Cleveland Clinic

Active smoking is not associated with severity of coronavirus disease 2019 (COVID-19)

Active smoking is associated with severity of coronavirus disease 2019 (COVID-19): An update of a meta-analysis

Association of smoking history with severe and critical outcomes in COVID-19 patients: A systemic review and meta-analysis

Smoking and risk of negative outcomes among COVID-19 patients: A systematic review and metaanalysis

Prevalence and underdiagnosis of chronic obstructive pulmonary disease among patients at risk in primary care

COPD Underdiagnosis and Misdiagnosis in a High-Risk Primary Care Population in Four Latin American Countries. A Key to Enhance Disease Diagnosis: The PUMA Study

Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for Diagnosis, Management and Prevention of COPD 2020

A c c e p t e d M a n u s c r i p t 15