key: cord-0745927-kt0euvx9
authors: Li, Dongmei; Croft, Daniel P.; Ossip, Deborah J.; Xie, Zidian
title: The Association between Statewide Vaping Prevalence and COVID-19
date: 2020-11-25
journal: Prev Med Rep
DOI: 10.1016/j.pmedr.2020.101254
sha: 248c2c0c89f7b4d0ef4c774248a81eeda8489953
doc_id: 745927
cord_uid: kt0euvx9

Existing literature indicated electronic cigarette users (vapers) have impaired immune response that might increase vulnerability to coronavirus disease 2019 (COVID-19) infection and death. However, whether vapers are more susceptible to COVID-19 is unknown. Using integrated data in each US state from the 2018 Behavioral Risk Factor Surveillance System (BRFSS), United States Census Bureauand the 1Point3Acres.com website, generalized estimating equation (GEE) models with negative binomial distribution assumption and log link functions were used to examine the association of statewide e-cigarette use prevalence with number of COVID-19 cases and deaths in the US on a state level from January 21, 2020 to April 25, 2020. The weighted proportion of vapers who used e-cigarettes every day or some days ranged from 2.86% to 6.42% for US states. Statistically significant associations were observed between the weighted proportion of vapers and number of COVID-19 cases as well as COVID-19 deaths in the US after adjusting for the weighted proportion of smokers and other significant covariates in the GEE models. With every one percent increase in weighted proportion of vapers in each state, the number of COVID-19 cases increase by 0.3139 (95% CI: 0.0554 – 0.5723) and the number of COVID-19 deaths increase by 0.3730 (95% CI: 0.0815 – 0.6646) in log scale in each US state. The positive associations between the proportion of vapers and the number of COVID-19 cases and deaths in each US state in this ecological study suggest an increased susceptibility of vapers to COVID-19 on a state level and warrants further investigation.

Novel coronavirus disease 2019 (COVID-19) outbreak was declared a global pandemic by the World Health Organization (WHO) on March 11, 2020. 1 As of April 28, 2020, there were over three million COVID-19 cases and over 200,000 deaths globally. 2 In the United States, the total number of COVID-19 cases exceeded one million, with over 57,000 deaths reported by April 28, 2020 . COVID-19 presents with cough, dyspnea and fever among other systemic symptoms and can lead to pneumonia and acute hypoxemic respiratory failure. 3 Electronic cigarettes (e-cigarettes), promoted as an alternative for cigarette smoking, rapidly gained popularity in recent years in the US. In 2018, the prevalence of current e-cigarette use (vaping) in US adults was 3.2%. 4 Recent studies on the associations of vaping and health symptoms/diseases have observed associations between vaping and symptoms of wheezing and self-reported Chronic Obstructive Pulmonary Disease (COPD), along with increased inflammation in bronchial epithelial cells and alterations in the pulmonary immune response to infection. [5] [6] [7] [8] [9] [10] Tobacco control researchers have raised concerns that vapers may be more susceptible to COVID-19 and could develop more severe COVID-19 symptoms. 11 However, there is very limited evidence on the association between vaping and COVID-19.

We examined the association of vaping with COVID-19 cases and deaths, using the integrated state-level weighted proportions of current e-cigarette users (vapers) from the 2018 Behavioral Risk Factor Surveillance System (BRFSS) survey data, the population size and land area in 2018 in each state from United States Census Bureau, and the daily number of COVID-19 cases and deaths in each state from the 1Point3Acres.com website during the time period from January 21, 2020 to April 25, 2020 in the United States (https://coronavirus.1point3acres.com). Our study is the first one to observe the association of vaping with COVID-19 at the US population level.

We integrated the 2018 Behavioral Risk Factor Surveillance System (BRFSS) survey data at state level (https://www.cdc.gov/brfss/annual_data/annual_2018.html), the population size and land area in each state from the United States Census Bureau COVID-19 infected cases and deaths counts were available for each state from January 21, 2020 to April 25, 2020. Reports of negative numbers of infected cases and deaths were excluded from the COVID-19 data. After integrating the BRFSS data and the census data with the COVID-19 cases and deaths from different dates at the state level, there were 1607 observations in the final analysis data.

The vaping status variable was defined by the answers to the question "Do you now use ecigarettes, every day, some days, or not at all?" in the 2018 BRFSS survey. Subjects who now use e-cigarettes every day or some days were classified as vapers and subjects who responded that they use e-cigarettes "not at all" or "not applicable" were classified as non-vapers. The weighted frequency of vapers in each state was obtained using the proc surveyfreq procedure in SAS version 9.4 (SAS Institute Inc., Cary, NC), considering the complex sampling design of the BRFSS survey. The weighted proportion of vapers in each US state was calculated using the ratio of weighted frequency of vapers and weighted frequency of total number of subjects in each state.

The outcomes used in current analysis were the number of COVID-19 cases and deaths in US states. Covariates considered in the current investigation included population size, population density (calculated using population size divided by land area) from the United States Census Bureau, age, gender, race/ethnicity, education, income, mental health, physical health, obesity, respiratory disease (including asthma and COPD), heart disease, cancer, stroke, diabetes, kidney disease, and smoking (currently smoke every day or some days) from the 2018 BRFSS survey data. The number of COVID-19 cases was also used as a covariate when modeling the COVID-19 deaths. All the covariates were state-level variables. The covariates included in the GEE model for COVID-19 cases were state population size in log scale, male, less than high school education, poor physical health, cancer, obese, and smoker. The covariates in the GEE model for COVID-19 deaths included the number of daily confirmed cases in each state, state population size in log scale, proportion of people aged 35-44, proportion of White, Black, and Hispanic, proportion of people having respiratory disease, cancer, being obese, and being smokers.

Generalized estimating equation (GEE) models with negative binomial distribution assumptions and log link functions were used to examine the association of weighted proportion of vapers with the number of COVID-19 cases and deaths, after adjusting for the confounding effects from significant covariates. 12, 13 The correlations of number of COVID-19 cases and deaths from different dates within the same state were considered through the autoregressive 1 (AR (1)) variance-covariance structure within the GEE model framework. The purposeful covariates selection method was used to select significant covariates for the GEE models. 14 Variance inflation factor (VIF) was used to examine the multicollinearities among the predictor variables in the GEE models. 15 A VIF value of five or less was considered to indicate no multicollinearity in the fitted GEE model. All statistical analyses were conducted using statistical analysis software SAS version 9.4 (SAS Institute Inc., Cary, NC) and R (R Core Team, 2017).

Significance levels for all tests were set at 5% for two-sided tests.

The weighted proportion of vapers ranged from 2.86% to 6.42% for US states. The daily number of infected COVID-19 cases ranged from 0 to 11,743 with an average of 362 daily cases and a median of 69 daily cases across all states in the US during the time period from January 21, 2020 to April 25, 2020. Figure 

Using integrated state level data obtained from the 2018 BRFSS survey, the United States Census Bureau and the 1Point3Acres.com website, we were able to investigate the association of weighted proportion of vapers with COVID-19 cases and deaths in the United States. We found a significant positive association of vaping prevalence with COVID-19 cases and deaths at the state level controlling for sociodemographic and health related covariates. This finding further supports the need for future research into the potential susceptibility of vapers to COVID-19.

Additional findings include the expected positive association between population size and COVID-19 cases and death. Finally we also observed a positive association between a low education level and COVID-19 cases and a negative association between male sex and COVID-19 cases, as well as negative associations of obesity with COVID-19 cases and death.

Though we identified a positive association between the weighted proportion of vapers and the number of COVID-19 cases and deaths at state level, we did not have data on what proportion of those who actually contracted COVID-19 or died from COVID-19 were vapers. Thus, we cannot determine causality between vaping and COVID-19 cases and deaths. However, prior research supports the biological plausibility of a relationship between vaping and an increased susceptibility to respiratory infection. 16 Multiple mouse models have observed an increased severity in respiratory infection associated with vaping exposure related to dysregulation of lung epithelial cells and an impaired immune response to both viral 17 , and bacterial infection. 18 Bacterial superinfection of viral illnesses like influenza and COVID-19 is especially dangerous, as this leads to an increased severity in illness 19 A human cell based model of exposure to nicotine-free flavored e-liquid observed immunosuppressive effects and impaired respiratory innate immune cell function (alveolar macrophages, neutrophils, and natural killer cells). 20 In humans, bronchoalveolar lavage samples from the airways of active vapers also revealed dysregulation of the airway's innate immune response including neutrophilic response and mucin. 21 A wide variety of flavorings are used by vapers, many of which, such as diacetyl, acetoin, pentanedione, o-vanillin, maltol, and coumarin in nicotine-free e-liquid, could also trigger inflammatory responses in human monocytes. 7 Previous human studies found vaping is associated with increased risk of chronic bronchitis symptoms (chronic cough or phlegm) [22] [23] [24] and epidemiologic studies observed an increased risk of self-reported wheezing and COPD. 9, 10 In addition, social behaviors associated with vaping could also enhance the risk of infection. For example, vapers who were not wearing a mask inside confined areas or even outside in a nondistanced group would be at an increased risk for contracting COVID-19. With the ongoing COVID-19 pandemic, particular health concerns have been raised regarding vaping, such as whether vapers have higher risk for COVID-19 and could develop more severe symptoms once contracted COVID-19. 11 To our best knowledge, this is the first ecological study to empirically examine and observe an association between vaping and COVID-19.

The existing literature on the increased risk of respiratory infection in combustible cigarette smokers was summarized by a prior meta-analysis finding an increased risk of current smokers for influenza infection compared to non-smokers. 25 As COVID-19 is a novel condition, the literature examining the risk of smokers for COVID-19 is scant. A recent study based on 1099 COVID-19 patients found smoking history was associated with COVID-19 severity. 26 A recent systematic review on COVID-19 and smoking concludes that smoking is likely associated with worse outcomes in COVID-19. 27 However, other studies indicate that smoking might not be associated with the incidence and severity of COVID-19 28 , including a recent meta-analysis based on Chinese patients suggests that active smoking is not associated with severity of COVID-19. 29 It remains unclear whether nicotine has a role in the either the increased or decreased severity of illness for smokers with COVID-19. Our study did not find a significant association between the weighted proportion of smokers and the number of COVID-19 infections and deaths at state level. Due to the incomplete testing and tracking of home deaths, it is possible that a percentage of older smokers with comorbidities are dying at home from COVID-19 and therefore are not captured into the reported COVID-19 cases and death data 30, 31 Additionally, it is unknown whether or not COVID-19 could be transmitted to those surrounding smokers through passive smoking and vaping. More epidemiological and clinical studies are needed to investigate the association of smoking with COVID-19 cases and deaths.

We found states that had a larger weighted proportion of subjects who had less than high school education had a higher number of COVID-19 cases. Researchers from the University of Southern California found that Americans who had less than high school education had a lower perceived risk of exposure to COVID-19 and a higher perceived risk of deaths than those who have college or higher degrees. 32 This might explain, in part, the positive association between the weighted proportion of less than high school education with the number of COVID-19 cases. We also found states that had a larger proportion of non-Hispanic Blacks and Hispanics had a larger number of COVID-19 deaths. The COVID-19 deaths rate data from Washington D.C. and 36 US states reported through April 27, 2020 showed that non-Hispanic Blacks (28.4%) and Hispanics (11.3%) had the highest COVID-19 deaths rate. 33 This could be related to the higher proportion of chronic conditions such as hypertension, heart disease and diabetes in non-Hispanic Blacks and Hispanics. 34, 35 In addition, COVID-19 deaths could also be affected by availability and access to care as well as the level of care, which could contribute to the ethnic disparities in COVID-19 deaths. Interestingly, we found that states with a higher proportion of males had lower number of COVID-19 cases than states with higher proportion of females. Previous investigation in Chinese COVID-19 patients found a larger proportion of males admitted to hospital and males also had more severe symptoms and higher mortality rate 26, [36] [37] [38] . This observation of a gender difference in COVID-19 cases and death and the role of gender in COVID-19 cases and death in the US is worth further investigation. In addition, we observed that states with a higher proportion of obesity had reduced COVID-19 cases and death. One possible explanation could be that people who are obese have a decreased exposure to COVID-19 due to being relatively home-bound due to disability or mobility issues. Although recent studies indicate obesity might be a risk factor for severe COVID-19 symptoms, the evidence base detailing the association of obesity with COVID-19 cases and death is still limited and need further investigation 39 .

There are several limitations in current study. First, the weighted proportions of vapers, smokers, and other demographic and chronic diseases are from the 2018 BRFSS data, which might differ from the 2020 estimates. The outbreak of e-cigarette, or vaping product use-associated lung injury (EVALI) occurred in September 2019 might change the vaping pattern on state levels as CDC issued guidance against the use of e-cigarettes. The reported COVID-19 cases and deaths obtained from 1Point3Acres.com website could be subject to some reporting errors as we noticed some negative number of COVID-19 cases and deaths, which we excluded from further analysis.

However, we compared the COVID-19 data obtained from 1Point3Acres.com website with the COVID-19 data from the New York Times 40 and Centers for Disease Control and Prevention (CDC) website 41 and found consistent numbers of COVID-19 cases and deaths, which increased the robustness and reliability of the data sources. Second, our current analysis is on state level instead of individual level. We don't know the individual status of vaping and COVID-19 cases or deaths, thus estimated coefficients of association could be different from epidemiological or clinical studies on individual subjects. Third, this ecological study at a group-level is also subject to limitations such as ecological bias, confounding, and misclassifications. The group-level data did not include information on whether vapers were strictly following social/physical distancing guidelines to reduce their risks for COVID-19. Fourth, the study period did not cover the whole duration in the US as the pandemic is still ongoing. The analysis of the daily number of cases and deaths over the first period of the pandemic could be biased as the spread of the virus was unequally across states in the US during the first three months and our ecological study was not able to account for outbreak dynamics (e.g. size of outbreak, public health guidelines). Finally, the number of reported cases could be impacted by testing strategies and reporting system that might be different from one state to another. 

The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the FDA.

The authors declare no conflicting interests.

The Prevalence of e-ciga.. 

WHO Declares COVID-19 a Pandemic

An interactive web-based dashboard to track COVID-19 in real time

Current Status of Epidemiology, Diagnosis, Therapeutics, and Vaccines for Novel Coronavirus Disease 2019 (COVID-19)

Prevalence of e-Cigarette Use Among Adults in the United States

Vaping Away Epithelial Integrity

Electronic cigarette vapour increases virulence and inflammatory potential of respiratory pathogens

E-cigarette flavored pods induce inflammation, epithelial barrier dysfunction, and DNA damage in lung epithelial cells and monocytes

Electronic cigarette liquid increases inflammation and virus infection in primary human airway epithelial cells

Association of smoking and electronic cigarette use with wheezing and related respiratory symptoms in adults: cross-sectional results from the Population Assessment of Tobacco and Health (PATH) study, wave 2

Use of Electronic Cigarettes and Self-Reported Chronic Obstructive Pulmonary Disease Diagnosis in Adults

Public health concerns and unsubstantiated claims at the intersection of vaping and COVID-19

A generalized estimating equations approach to mixed-effects ordinal probit models

Modelling correlated data: Multilevel models and generalized estimating equations and their use with data from research in developmental disabilities

Purposeful selection of variables in logistic regression

Applied Linear-Regression Models -Neter

Public Health Consequences of E-Cigarettes

Electronic cigarettes disrupt lung lipid homeostasis and innate immunity independent of nicotine

Electronic cigarette inhalation alters innate immunity and airway cytokines while increasing the virulence of colonizing bacteria

Influenza and Bacterial Superinfection: Illuminating the Immunologic Mechanisms of Disease

Flavored e-cigarette liquids and cinnamaldehyde impair respiratory innate immune cell function

E-Cigarette Use Causes a Unique Innate Immune Response in the Lung, Involving Increased Neutrophilic Activation and Altered Mucin Secretion

What are the respiratory effects of ecigarettes?

Electronic Cigarette Use and Respiratory Symptoms in Chinese Adolescents in Hong Kong

Electronic Cigarette Use and Respiratory Symptoms in Adolescents

Cigarette smoking and the occurrence of influenza -Systematic review

Clinical Characteristics of Coronavirus Disease 2019 in China

COVID-19 and smoking: A systematic review of the evidence

Sex difference and smoking predisposition in patients with COVID-19

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

What is happening to non-covid deaths

The association of smoking status with SARS-CoV-2 infection, hospitalisation and mortality from COVID-19: A living rapid evidence review

Racial/ethnic disparities in prevalence, treatment, and control of hypertension among US adults following application of the 2017 American College of Cardiology/American Heart Association guideline

Age-stratified prevalence, treatment status, and associated factors of hypertension among US adults following application of the 2017 ACC/AHA guideline

Epidemiology Working Group for Ncip Epidemic Response CCfDC, Prevention. [The epidemiological characteristics of an outbreak of 2019 novel coronavirus diseases (COVID-19)

Gender Differences in Patients With COVID-19: Focus on Severity and Mortality

Clinical Characteristics of Patients Who Died of Coronavirus Disease 2019 in China

Prevalence of obesity among adult inpatients with COVID-19 in France. The Lancet Diabetes and Endocrinology

DL, ZX: conceived and designed the study. DL: analyzed the data. DL wrote the manuscript. DL, DC, DO, ZX: assisted with interpretation of analyses and edited the manuscript.