key: cord-0955395-m67m3g4v
authors: Puebla Neira, Daniel; Watts, Abigail; Seashore, Justin; Polychronopoulou, Efstathia; Kuo, Yong-Fang; Sharma, Gulshan
title: Smoking and risk of COVID-19 hospitalization
date: 2021-04-17
journal: Respir Med
DOI: 10.1016/j.rmed.2021.106414
sha: cda3c09dc64cf10477832fcb207c7601ef98a042
doc_id: 955395
cord_uid: m67m3g4v

RATIONALE: The association between smoking status and severe Coronavirus Disease 2019 (COVID-19) remains controversial. OBJECTIVE: To assess the risk of hospitalization (as a marker of severe COVID-19) in patients by smoking status: former, current and never smokers, who tested positive for the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV2) at an academic medical center in the United States. METHODS: We conducted a retrospective cohort study in patients with SARS-COV2 between March-1-2020 and January-31-2021 to identify the risk of hospitalization due to COVID-19 by smoking status. RESULTS: We identified 10216 SARS-COV2-positive patients with complete documentation of smoking habits. Within 14 days of a SARS-COV2 positive test, 1150 (11.2%) patients were admitted and 188 (1.8%) died. Significantly more former smokers were hospitalized from COVID-19 than current or never smokers (21.2% former smokers; 7.3% current smokers; 10.4% never smokers, p<0.0001). In univariable analysis, former smokers had higher odds of hospitalization from COVID-19 than never smokers (OR 2.31; 95% CI 1.94-2.74). This association remained significant when analysis was adjusted for age, race and gender (OR 1.28; 95% CI 1.06-1.55), but became non-significant when analysis included Body Mass Index, previous hospitalization and number of comorbidities (OR 1.05; 95% CI 0.86-1.29). In contrast, current smokers were less likely than never smokers to be hospitalized due to COVID-19. CONCLUSIONS: Significantly more former smokers were hospitalized and died from COVID-19 than current or never smokers. This effect is mediated via age and comorbidities in former smokers.

Since first being recognized in December 2019, the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV2) has caused to date 130 million cases and 2.8 million deaths worldwide 1 . Coronavirus Disease 2019 , the infectious disease caused by SARS-COV2, can range in presentation from asymptomatic to severe disease. When symptomatic, the time from start of illness to when patients experience dyspnea ranges from 5 to 8 days. About 3-15% of individuals infected require hospitalization [2] [3] [4] [5] . In patients who develop severe disease the time from the beginning of illness to when patients develop acute respiratory distress syndrome (ARDS) ranges from 8 to 12 days. Also, the time from onset of illness to Intensive Care Unit (ICU) admission ranges from 9.5 to 12 days 6 . Risk factors for severe and greater risk of mortality have been identified 5, [7] [8] [9] ; specifically, these are: age greater than 65 years, coronary artery disease (CAD), congestive heart failure (CHF), cardiac arrhythmia, chronic obstructive pulmonary disease (COPD), diabetes mellitus, cancer, and chronic kidney disease (CKD), obesity, black race and Hispanic ethnicity. 5, [8] [9] [10] [11] [12] [13] .

Despite agreement about the association between severe COVID-19 and the comorbidities mentioned above, the relationship between tobacco use and the severity of COVID-19 infection remains controversial 8,10,12,14-20 . Smoking has been associated with increased risk of infection and worse outcomes for a multitude of bacterial and viral pathogens 21 . For instance, smokers with or without chronic lung disease have higher risk of invasive pneumococcal disease and community acquired pneumonia 22, 23 . Also, smoking is a risk factor for the development of and death from active tuberculosis infection [24] [25] [26] . Similarly, smokers are more likely to be infected J o u r n a l P r e -p r o o f with and develop clinical symptoms when exposed to rhinovirus, respiratory syncytial virus or coronavirus type 229 27 . Influenza infection and severity are worse in smokers compared to non-smokers [28] [29] [30] . In contrast, studies throughout the world have reported a low prevalence of current smokers among people that tested positive for SARS-COV2 and/or were hospitalized due to COVID-19 10, 12, [18] [19] [20] . The association between smoking and severe COVID-19 is unknown.

To evaluate the relationship between former, current, and never smoker status and severe COVID-19, we conducted a retrospective cohort study to identify the risk of hospitalization from COVID-19 (as a marker of severe infection) within 14 days of a positive SARS-COV2 test in former and current smokers compared to never smokers.

Based on the known adverse effects of smoking, we hypothesize that smokers (former and/or current) may have higher risk of hospitalization from COVID-19 compared to never smokers.

In this retrospective cohort study patients at the University of Texas Medical Branch (UTMB) Health System with any SARS-COV2 positive laboratory test result from March 1, 2020 to January 31, 2021 were included. The UTMB institutional review board approved the study (IRB# 20-0249). Data was collected from Epic, the institution's electronic health records system (EHR). Patients who were prisoners, younger than 18 J o u r n a l P r e -p r o o f years, admitted after 14 days of an initial positive SARS-COV2 result, had no clinical encounter in the prior 2 years in our health system or had incomplete smoking status records in the 2 years before SARS-COV2 test were excluded ( Figure 1 ). We divided the study cohort into former, current and never smokers based on each patient's lifetime count of cigarettes smoked and most recent self-reported smoking status. Patients who self-reported as a never smoker or who had smoked less than 100 cigarettes in their lifetime were considered never smokers; patients who smoked more than 100 cigarettes in their lifetime were considered smokers (former or current ) 31 . Former smokers' status was assigned based on the patient's self-report of "former smoker" in the EHR.

Our main outcome was any COVID-19 related hospitalization within 14 days of the initial SARS-COV2-positive result. COVID-19-related hospitalization was defined as COVID- 

Patient characteristics were summarized with mean, standard deviation (SD), frequency and percentage. In univariable analysis, we compared former, current and never smokers. For categorical characteristics, we used Chi-square test and Fisher's exact test; for continuous characteristics, we used analysis of variance. Multivariable analyses exploring the relationship of smoking status with the outcome were performed by fitting multivariable logistic regression models, adjusted for age, gender, race, BMI, number of comorbidities and number of inpatient stays within the previous year. The interaction of smoking status with other characteristics was evaluated. As sensitivity analysis, we split our cohort in three groups based on age (18-40, 41-64 and 65 years or older). Then, we fitted multivariable logistic regression models separately for each age group to explore any effect of smoking status on the risk of COVID-19 hospitalization. We also conducted sensitivity analyses with different BMI cutoffs, added Hispanic ethnicity and analyzed age as a continuous variable. Data management was performed with Python 3.7 and analysis was performed with SAS version 9.4 (SAS Inc., Cary, NC). All reported pvalues were two-sided with p<0.05.

Demographics Detailed characteristics of the cohort are presented in Table 1 

We found 1,150 (11.2%) patients who were hospitalized due to COVID-19 within 14 days of a positive test for SARS-COV2. Significantly more former smokers were hospitalized from COVID-19 than current or never smokers (former smokers, 21.2%; current smokers, 7.3%; never smokers, 10.4%; p<0.0001). (Table 2 ). Univariable analysis showed that former smokers had significantly higher odds of hospitalization from COVID-19 compared to never smokers (Odds Ratio [OR] 2.31; 95% Confidence Interval [CI] 1.94-2.74). In contrast, current smokers were less likely to be hospitalized from COVID-19 than never smokers (OR 0.68; 95% CI 0.60-0.99). When analysis was adjusted for age, race and gender, former smokers' odds of hospitalization due to COVID-19 remained higher than that of never smokers (OR 1.28; 95% CI 1.06-1.55); and current smokers' odds of hospitalization remained lower than that of never smokers. During further analysis (adding BMI, previous inpatient hospitalizations and number of comorbidities), there was a trend towards increase in odds of hospitalization J o u r n a l P r e -p r o o f for former smokers compared to never smokers, but the difference did not reach statistical significance (OR 1.05; 95% CI 0.86-1.29). Current smokers' odds of hospitalization due to COVID-19 remained lower than that of never smokers (OR 0.55; 95% CI 0.37-0.83).

As sensitivity analysis, we stratified our cohort by age (18-40, 41-64, and 65 years or older) and adjusted for all other variables. Here we found an increasing trend for the association between being a former-smoker and hospitalization from COVID-19, but it did not reach statistical significance. Current smokers were less likely to be hospitalized due to COVID-19 compared to never smokers, but this was significant only in patients younger than 65 years ( Table 4) .

Age: The odds of hospitalization from COVID-19 within 14 days following a positive test for SARS-COV2 increased gradually for each age group studied, ranging from 3.23 for those 40-49 years (OR 3.23; 95% CI 2.51-4.14) to 44.82 (OR 44.82; 95% CI 31.05-64.69) for those aged 85 years or more, compared to those aged 40 years or less ( Comorbidities and prior inpatient admission: Patients with 3 or more comorbidities were more likely to be hospitalized than patients with fewer than 3 comorbidities (OR 1.45; 95% CI 1.21-1.73). The odds for hospitalization were more than 3 times higher in patients who had an inpatient admission in the prior year compared to those not admitted (OR 3.64; 95% CI 3.03-4.37).

Gender: In contrast to the above risk factors, women had 43% lower odds of admission compared to men (OR 0.57; 95% CI 0.5-0.66).

We found 188 patients (1.8%) who died within 14 days of a positive test for SARS-COV2. A significantly higher percentage of former smokers died within 14 days of J o u r n a l P r e -p r o o f positive testing for the virus than current or never smokers (former smokers 5.1%; current smokers 0.8%; never smokers 1.6%; p<0.0001) ( Table 2 ).

In our study, patients who were former smokers had a higher percent of hospitalizations and death within 14 days of a positive test for SARS-COV2. After adjustment for other covariates, there was a trend for increase odds of hospitalization in former smokers compared to never smokers, but the trend did not reach statistical significance. Current smokers had a lower percentage of hospitalizations and death than did former or never smokers. They were also less likely than never smokers to be hospitalized due to Our findings of a higher percentage of hospitalizations and death for severe in former smokers compared to never smokers is intriguing. The risk of severe COVID-19 among former smokers is significantly driven by the effect of age and comorbidities.

In our cohort, the mean age of former smokers was 10 years older than that of current smokers, and 12 years older than that of never smokers. Also, despite our initial analysis showing increased odds of hospitalization in former smokers compared to never smokers, when it was adjusted for age we found a modest reduction in odds of J o u r n a l P r e -p r o o f hospitalization due to COVID-19 in former smokers compared to never smokers. Furthermore, after the analysis was adjusted for comorbidities, the association was no longer significant. Smoking has been causally linked to stroke, coronary artery disease, chronic obstructive pulmonary disease, asthma, interstitial lung disease, diabetes, cancer, impaired immune function, overall diminished health and premature death 21, 35, 36 . The risks of smoking-related disease result largely from cumulative damage; hence, the consequences of smoking occur disproportionately among the elderly 21, 35, 37, 38 . In our study, we estimated that the odds of hospitalization from COVID-19 increased by 6% for every year of age in the population studied. Older age and more comorbidities as shown in our and others' analysis 5, [8] [9] [10] 12 are significantly associated with severe COVID-19. Both variables may be significant confounders in the retrospective search for an association between smoking and severe COVID-19.

Several mechanisms have been proposed to describe the association between tobacco use and severe COVID-19. Tobacco smoke impairs mucociliary clearance, weakens innate and adaptive immune responses 39 and increases the risk of both viral and bacterial pulmonary infections 21, 22, [24] [25] [26] [27] [28] [29] [30] 40 . Tobacco use has been shown to increase levels of angiotensin-converting enzyme 2 (ACE2), the unique receptor used by SARS-COV2 to enter host cells [41] [42] [43] .

Alternatively, a potentially protective role of tobacco use in COVID-19 infection has been theorized. Nicotine, a highly addictive, psychoactive alkaloid contained in tobacco products, has shown immunomodulatory and anti-inflammatory effects 44, 45 . Tobacco use has also been linked to increased nitric oxide production in the lungs, resulting in diminished viral replication and impaired viral entry into host cells 46 . Nicotine has also J o u r n a l P r e -p r o o f demonstrated a protective effect for ARDS animal models, and is thought to potentially inhibit Interleukin-6, a key player in the cytokine release syndrome induced by SARS-COV2 infection 38 . And while some studies report upregulation of ACE2 receptors for SARS-COV2 cell entry in tobacco users, others challenge this claim and report an interaction between nicotine and the renin-angiotensin system, resulting in reduced ACE2 levels 47 . Moreover, several series throughout the world have reported a low prevalence of current smokers among people that tested positive for SARS-COV2 and/or were hospitalized due to COVID-19 10,12,18-20 .

Our study has several limitations. We relied on patients' self-report of smoking habits and on identification of smokers (current and former) in our EHR; as shown in others' analysis, smoking history can be accidentally altered when providers update the fields for the "number of packs-smoked-per-day" and the "total-years-smoked" 48 . Also, we do not know exactly when patients stopped smoking, and for the definition of former smokers this may be significant. It is well known that a longer time of smoking cessation is associated with more health benefits to patients. For instance, patients who stop smoking before age 40 avoid more than 90% of the excess mortality caused by continuing smoking 49 . Despite these challenges, and to strengthen our analysis, we excluded patients with no complete documentation of smoking history. Finally, our retrospective analysis may only assume an association between smoking status (former and/or current) and risk of severe COVID 19, and not a cause and effect relation. 

In conclusion, the findings of our study showed that significantly more formers smokers were hospitalized and died from COVID-19 than current or never smokers. This effect is mediated through age and comorbidities in former smokers. -Former, current and never smokers categories were based on each patient's lifetime count of cigarettes smoked and most recent self-reported smoking status.

-Patients who self-reported as a never smoker or who had smoked less than 100 cigarettes in their lifetime were considered never smokers; patients who smoked more than 100 cigarettes in their lifetime were considered smokers (former or current ) 31 .

-Former smokers' status was assigned based on the patient's self-report of "former smoker" in the EHR. Definition of Abbreviations: SARS-CoV2= Severe Acute Respiratory Distress Coronavirus 2. For patients with multiple records, we took the latest. A never smoker was defined as a patient who: 1) self-reported as never having been a smoker or 2) had <100 cigarettes in their lifetime. A current or former smoker was defined as a patient who self-reported as a current or former smoker and had > 100 cigarettes in their lifetime 31 . Current or former was based on latest self-report. Of patients who visited UTMB and had a positive test for SARS-COV2, 81.4% had smoking records. Of these patients who had a smoking record and fulfilled other eligibility criteria, 83.9% had complete records, including pack-per-day and smoking years and were included in the analysis. 

World Health Organization, WHO coronavirus disease (COVID-19) dashboard

Estimating the burden of SARS-CoV-2 in

Suppression of a SARS-CoV-2 outbreak in the Italian municipality of Vo'

Estimates of the severity of coronavirus disease 2019: a model-based analysis

Prevalence of underlying diseases in hospitalized patients with COVID-19: a systematic review and metaanalysis

Ten clinical tips on COVID-19 for healthcare providers involved in patient care

Characterization of the inflammatory response to severe COVID-19 illness

The impact of COPD and smoking history on the severity of COVID-19: a systemic review and meta-analysis

COVID-19 with different severities: a multicenter study of clinical features

Clinical characteristics of coronavirus disease 2019 in China

Outcomes of COVID-19: disparities in obesity and by ethnicity/race

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Association of race and ethnicity with COVID-19 test positivity and hospitalization is mediated by socioeconomic factors

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

Reply to: "Current smoking is not associated with COVID-19

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

Medical vulnerability of young adults to severe COVID-19 illness-data from the National Health Interview Survey

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

Presenting characteristics, comorbidities, and outcomes among 5700 patients hospitalized with COVID-19 in the New York City area

Renin-Angiotensin-Aldosterone System Inhibitors and Risk of Covid-19

Cigarette smoking and infection

Cigarette smoking and invasive pneumococcal disease, Active Bacterial Core Surveillance Team

XBalanzó, IBolíbar, Proportion of community-acquired pneumonia cases attributable to tobacco smoking

Risk factors associated with the prevalence of pulmonary tuberculosis among sanitary workers in Shanghai

Cigarette smoking as a risk factor for tuberculosis in young adults: a case-control study

Smoking and mortality from tuberculosis and other diseases in India: retrospective study of 43000 adult male deaths and 35000 controls

Smoking, alcohol consumption, and susceptibility to the common cold

Cigarette smoking and epidemic influenza

Smoking and epidemic influenza-like illness in female military recruits: a brief survey

Cigarette smoking as a risk factor for epidemic a(h1n1) influenza in young men

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

Low rate of daily active tobacco smoking in patients with symptomatic COVID-19

Smoking, vaping and hospitalization for COVID-19

National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health, The Health Consequences of Smoking-50 Years of Progress: A Report of the Surgeon General

The influence of cigarette smoking on viral infections: translating bench science to impact COPD pathogenesis and acute exacerbations of COPD clinically

Cigarette smoking among the elderly: disease consequences and the benefits of cessation

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

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

Tuberculosis risk factors in adults in

Cigarette smoking and COVID-19: a complex interaction

Tobacco smoking increases the lung gene expression of ACE2, the receptor of SARS-CoV-2

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

Nicotine exerts an anti-inflammatory effect in a murine model of acute lung injury

COVID-19: counter-intuitive data on smoking prevalence and therapeutic implications for nicotine

Nitric oxide inhibits the replication cycle of severe acute respiratory syndrome coronavirus

Nicotine and the renin-angiotensin system

The impact of fluctuations in pack-year smoking history in the electronic health record on lung cancer screening practices

The 21st century hazards of smoking and benefits of stopping: a prospective study of one million women in the UK

Daniel Puebla Neira, MD accepts full responsibility for the content of the manuscript including the data and analysis.The authors acknowledge and greatly appreciate the assistance in the preparation of this manuscript by: 

Other specified health status Z99.81Dependence on supplemental oxygen -ICD-10 codes above reflect the reason the patient sought care; not the discharge diagnosis. Many patients that presented with nonspecific complaints (i.e., fatigue or generalized weakness) eventually were admitted for hypoxia or respiratory failure and were found to be positive for SARS-COV2. Definition of Abbreviation: ICD= International Classification of Diseases; COVID-19= Coronavirus Disease 2019.