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

Importance: As the pandemic of COVID-19 is still under progression, identification of prognostic factors is a global challenge. Among epidemiological risk factors, the role of smoking, to date, is unclear. Objective: T o accurately evaluate a possible correlation of daily smoking with an increased susceptibility to SARS-CoV-2 infection. Participants: We estimated the rates of daily current smokers in COVID-19-infected patients in a large French university hospital between February 28th 2020 and March 30th 2020 for outpatients and from March 23rd till April 9th 2020 for inpatients. Design: T he rates from both groups were compared to those of daily current smokers in the 2018 French general population, established in 2018, after standardization of the data for sex and age. Results: T he inpatient group was composed of 343 patients, median age 65 yr: 206 men (60.1%, median age 66 yr) and 137 women (39.9%, median age 65 yr) with a rate of daily smokers of 4.4% (5.4% of men and 2.9% of women).T he outpatient group was composed of 139 patients, median age 44 yr: 62 men (44.6 %, median age 43 yr, and 77 women (55.4 %, median age 44). T he daily smokers rate was 5.3% (5.1% of men and 5.5 % of women). In the French population, the daily smokers rate is 25.4% (28.2% of men and 22.9% of women). T he rate of current daily smokers was significantly lower in COVID-19 outpatients and inpatients (80.3% and 75.4%, respectively), as compared to that in the French general population with standardized incidence ratios according to sex and age of 0.197 [0.094 0.41] and 0.246 [0.148 0.408]. T hese ratios did not significantly differ between the two groups (P=0.63). Conclusions and relevance: Our study highlights for the first time a dramatic decreased prevalence of daily current smokers both in COVID-19 outpatients and inpatients as compared to the general population. Qeios, CC-BY 4.0 · Article, April 19, 2020 Qeios ID: WPP19W · https://doi.org/10.32388/WPP19W 1/16 In t rodu ct i on As the pandemic of COVID-19, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is still under progression, the identification of prognostic factors is a global challenge. Among epidemiological risk factors, the role of smoking, to date, is unclear. Smoking has been initially found associated with adverse disease prognosis of COVID-19[1], although this finding remains controversial[2]. Reported rates of current smokers among SARS-CoV-2-infected patients are heterogeneous, ranging from 1.4% to 12.5% (T ABLE 1[1, 3-10]). T he rates of current smokers remain however strikingly low for the middle-aged Chinese population (median age: 47.0 years; range: 35.0–58.0; in Guan et al.[1]). T hese data notwithstanding, no firm conclusions can be drawn from the available COVID-19 studies because main potential confounders, like age and sex, were not taken into account. Additionally, these studies included mostly hospitalized patients, and the low rate of current smokers may be related to high rate of patients with comorbidities (smokers having been advised to quit) and thus to COVID-19 severity. T his could therefore introduce a confusion bias. T o allow for a valid comparison with the general population, smoking rates used as a reference should have been evaluated at a time as close as possible to time of COVID pandemic and the same definitions of current smokers should be used for both the COVID-19 population and the general population, which was not clear in the previous studies. T he last available study of the Chinese general population in 2015 reported rates of current smokers of 52% for men and 2.5 % for women[11]. Very recently, the US Center of Disease Control reported an analysis of current smoker rate among US COVID-19 patients which was found to be 1.3% for the whole population of COVID-19 patients, 1% for outpatients, 2% for patients, not hospitalized in an ICU, and 1% in ICU-admitted patients[12]. However, the level of missing smoking status was very high and no comparison with the general population was performed. T herefore, the hypothetic « protective » effect of current smoking on the risk of SARSCoV-2 infection that can be extrapolated from the low current smoker rate has yet to be determined. T o accurately evaluate whether or not current smoking is associated with the risk of contracting a symptomatic SARS-CoV-2 infection, we have compared the rates of current smokers of two groups, one composed of both COVID-19 outpatients (not subsequently hospitalized) and one of COVID-19 inpatients with those in the 2018 French general population after standardization by sex and age.[13] Qeios, CC-BY 4.0 · Article, April 19, 2020 Qeios ID: WPP19W · https://doi.org/10.32388/WPP19W 2/16 N (total number of patients) Median Sex(%) Country % of current smokers Current Possible confounders of smoking status analysis Ref Age (yr) Male/female Smoking rate (%) in the general population Male/Female (date) 1099 47 58.1/41.9 China 12.6 52%/ 2.5 % (2015) Outpatients and Inpatients mixed 1 Age, and sex not accounted 191 56.0 62/38 China 6.0 “ Inpatients only 10 No age, and sex –matched control population 41 49 73/27 China 7.0 “ Inpatients only 3 No age, and sex –matched control population 52 59.7 * 67/33 China 4.0 “ Inpatients only 8 No age, and sex –matched control population 140 57 50.7/49.3 China 1.4 “ Inpatients only 9 No age, and sex –matched control Table 1 Prevalence of smokers in COVID-19 patients in published series Mat er i a l an d met h ods Pat i ent s and d es i g n T his is a cross-sectional survey investigating the rate of current smokers in patients with a diagnosis of COVID-19, both in hospitalized patients (representing the severe symptomatic cases of COVID-19) and in outpatients (i.e. patients who represent the nonsevere symptomatic cases of this infection). Current smoker rates were compared to those of the French population as a reference, after standardization by age and sex. Eligible patients were those with a confirmed diagnosis of COVID-19 at the APHP PitiéQeios, CC-BY 4.0 · Article, April 19, 2020 Qeios ID: WPP19W · https://doi.org/10.32388/WPP19W 3/16 Salpêtrière Hospital, Paris, France, either hospitalized in medical wards of medicine, but not in ICUs (inpatients) or having consulted for this infection in the infectious disease department and who did not require hospital care until end of the acute infectious episode (outpatients). T he period of inclusion ran from the 23rd of March to 9th of April 2020 for inpatients and from the 28th of February to the 30th of March 2020 for outpatients. T his study is observational. All data were collected in the context of care and were collected in completely anonymous sheets and therefore, in accordance with the French law, including the GPRD, informed consent of the patient was not sought. T he study has been approved by the ethics committee of the Sorbonne University (2020 CER-202013) Def i n i t i ons and d at a c o l l ec t ed Confirmed COVID-19 was defined as a positive result on real-time reverse-transcriptase– polymerase-chain-reaction (RT -PCR) assay of nasal and pharyngeal swab specimens. Smoking status was collected and patients were specifically asked whether they were current smokers (and if so, to provide details on their smoking habits: daily or occasional smoking, type of tobacco products used, number of daily cigarettes), former smokers, or not smokers ever). Daily smokers are individuals reporting daily smoking or reporting a daily frequency of the number of cigarettes (manufactured or rolled) or other tobacco products (cigars, cigarillos, pipe, shisha)[13]. Occasional smokers are individuals reporting infrequent, but not daily smoking. T he group of ex-smokers includes anyone having smoked in the past, occasionally or daily, and had abstained from smoking prior to the time of investigation. T he term "never smoker" designated people who had never smoked. T he quantities of tobacco smoked were calculated using the following equivalences: 1 cigar = 1 cigarillo = 2 cigarettes. In addition to smoking status, the following data were extracted from the medical charts: age, sex, comorbidities, known to have potentially an impact on the prognosis of COVID19, including diabetes, hypertension, obesity, immunodepression and respiratory disease (such as COPD, other clinical manifestations of COVID-19 and outor inpatient status. Sm ok i ng r at e i n t he p op ul at i on of r ef er enc e T he French general population was used as a reference to compute the SIR. Recent rates of current daily smokers have been reported for the year 2018 by sex and age class (of 10 years) from the General Survey “Baromètre de Santé Publique France” of the French population, cross sectional phone survey made yearly on a representative sample of 18Qeios, CC-BY 4.0 · Article, April 19, 2020 Qeios ID: WPP19W · https://doi.org/10.32388/WPP19W 4/16 75 year-old people living in mainland France, with a on 2-level random sampling[13]. T he 2018 survey involved a sample of 9,074 individuals. T he completion of the survey took place from January 10 to July 25, 2018 and used the same definitions of daily smokers, occasional smokers, former smokers and never smokers as described above. St at i s t i c a l anal ys i s A descriptive analysis has been made by group (inpatients outpatients). Qualitative variables were described by numbers and percentages, and quantitative variables by median and interquartile range. Inpatients and outpatients were compared for qualitative variables with Pearson Chi2 tests or Fisher’s exact test as appropriate and for quantitative variables with Wilcoxon sum rank test. T he SIRs were used to compare current daily smoker rates in the COVID-19 inpatients and outpatients, respectively, with those of current daily smokers in a reference population, here the French general population in 2018. T he estimated SIR and its 95% confidence interval is the ratio between the observed number of current daily smokers among the COVID-19 patients and the number of current daily smokers that would be expected, on the basis of ageand gender-specific current daily smokers rates in the general population. T he main analysis involved all included patients, and those older than 75 years were considered in the 65-75 years age class for standardization, which for our hypothesis is conservative approach, because current smoker rates decreases with age. For 7 outpatients and 2 inpatients, medical charts, and thus smoking status, was not available. We did not include the latter patients in the main analysis for reasons that the lack of medical history was very likely to be at random regarding smoking status. We performed two sensitivity analysis, one excluding patients older than 75 years, the other considering these nine patients with missing data on the smoking status as current smokers. Resu l t s Dem og r ap hi c and Cl i n i c a l Char ac t er i s t i c s A total of 343 inpatients and 139 outpatients were included. T he demographic and clinical characteristics of the two groups are shown in T ABLE 2. T he inpatient group was composed of 343 patients, median age 65 yr: 206 men (60.1%, median age 66 yr) and 137 women (39.9%, median age 65 yr). T he rate of daily smokers was 4.4 (5.4% of men and 2.9% of women). T he outpatient group was composed of 139 patients, median age 44 yr: 62 men (44.6 %, Qeios, CC-BY 4.0 · Article, April 19, 2020 Qeios ID: WPP19W · https://doi.org/10.32388/WPP19W 5/16 median age 43 yr, and 77 women (55.4 %, median age 44). T he daily smokers rate was 5.3% (5.1% of men and 5.5 % of women). As described by others[1], hypertension (41.4%), diabetes (27.7%), obesity (14.4%) and immune deficiencies (17.8%) are frequently observed in inpatients while COPD is less frequent (7.9%). T hose comorbidities were significantly less frequent in outpatients with hypertension counting for 12.1%, (P<0.0001), diabetes for 5.3% (P<0.0001), obesity for 7.6% (P=0.045), immunodeficiencies for 3.0 % (P<0.0001), and COPD for 1.5% (P= 0.009). As shown in figure 1, age distribution differs between outpatients and inpatients with younger patients in outpatients and older patients in inpatients. Fig ure 1 Ag e pyra mid of COVID-19 inpa tie nts a nd outpa tie nts . Dark and light shaded histograms represent outpatients and inpatients with confirmed COVD-19 status, respectively T he rates of daily current smokers in inpatients (4.4%) did not significantly differ from that in outpatients (5.4%; P= 0.67; T ABLE 2). Occasional smoking was slightly more frequent in outpatients than in inpatients (4.6 vs 1.8 %; P=0.10) but the number of people concerned was small. Qeios, CC-BY 4.0 · Article, April 19, 2020 Qeios ID: WPP19W · https://doi.org/10.32388/WPP19W 6/16 Outpatients Inpatients Outpatient/inpatient comparison

I n t r od u ct i on I n t r od u ct i on As the pandemic of COVID-19, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is still under progression, the identification of prognostic factors is a global challenge. Among epidemiological risk factors, the role of smoking, to date, is unclear. Smoking has been initially found associated with adverse disease prognosis of COVID-19[1], although this finding remains controversial [2]. Reported rates of current smokers among SARS-CoV-2-infected patients are heterogeneous, ranging from 1.4% to 12.5% (T AB LE 1 T AB LE 1[1, [3][4][5][6][7][8][9][10]). T he rates of current smokers remain however strikingly low for the middle-aged Chinese population (median age: 47.0 years; range: 35.0-58.0; in Guan et al. [1]). T hese data notwithstanding, no firm conclusions can be drawn from the available COVID-19 studies because main potential confounders, like age and sex, were not taken into account. Additionally, these studies included mostly hospitalized patients, and the low rate of current smokers may be related to high rate of patients with comorbidities (smokers having been advised to quit) and thus to COVID-19 severity. T his could therefore introduce a confusion bias. T o allow for a valid comparison with the general population, smoking rates used as a reference should have been evaluated at a time as close as possible to time of COVID pandemic and the same definitions of current smokers should be used for both the COVID-19 population and the general population, which was not clear in the previous studies. T he last available study of the Chinese general population in 2015 reported rates of current smokers of 52% for men and 2.5 % for women [11]. Very recently, the US Center of Disease Control reported an analysis of current smoker rate among US COVID-19 patients which was found to be M a t e r i a l a n d m e t h od s M a t e r i a l a n d m e t h od s P a t i e n t s a n d d e s i g n P a t i e n t s a n d d e s i g n T his study is observational. All data were collected in the context of care and were collected in completely anonymous sheets and therefore, in accordance with the French law, including the GPRD, informed consent of the patient was not sought. T he study has been approved by the ethics committee of the Sorbonne University (2020 -CER-2020- Smoking status was collected and patients were specifically asked whether they were current smokers (and if so, to provide details on their smoking habits: daily or occasional smoking, type of tobacco products used, number of daily cigarettes), former smokers, or not smokers ever).
Daily smokers are individuals reporting daily smoking or reporting a daily frequency of the number of cigarettes (manufactured or rolled) or other tobacco products (cigars, cigarillos, pipe, shisha) [13]. Occasional smokers are individuals reporting infrequent, but not daily smoking. T he group of ex-smokers includes anyone having smoked in the past, occasionally or daily, and had abstained from smoking prior to the time of investigation.
T he term "never smoker" designated people who had never smoked. T he quantities of tobacco smoked were calculated using the following equivalences: 1 cigar = 1 cigarillo = 2 cigarettes.
In addition to smoking status, the following data were extracted from the medical charts: age, sex, comorbidities, known to have potentially an impact on the prognosis of COVID-19, including diabetes, hypertension, obesity, immunodepression and respiratory disease 2018 survey involved a sample of 9,074 individuals. T he completion of the survey took place from January 10 to July 25, 2018 and used the same definitions of daily smokers, occasional smokers, former smokers and never smokers as described above.
S t a t i s t i c a l a n a l y s i s S t a t i s t i c a l a n a l y s i s A descriptive analysis has been made by group (inpatients -outpatients). Qualitative variables were described by numbers and percentages, and quantitative variables by median and interquartile range. Inpatients and outpatients were compared for qualitative variables with Pearson Chi2 tests or Fisher's exact test as appropriate and for quantitative variables with Wilcoxon sum rank test.
T he SIRs were used to compare current daily smoker rates in the COVID-19 inpatients and outpatients, respectively, with those of current daily smokers in a reference population, here the French general population in 2018. T he estimated SIR and its 95% confidence interval is the ratio between the observed number of current daily smokers among the COVID-19 patients and the number of current daily smokers that would be expected, on the basis of age-and gender-specific current daily smokers rates in the general population. T he main analysis involved all included patients, and those older than 75 years were considered in the 65-75 years age class for standardization, which for our hypothesis is conservative approach, because current smoker rates decreases with age.
For 7 outpatients and 2 inpatients, medical charts, and thus smoking status, was not available. We did not include the latter patients in the main analysis for reasons that the lack of medical history was very likely to be at random regarding smoking status. We performed two sensitivity analysis, one excluding patients older than 75 years, the other considering these nine patients with missing data on the smoking status as current smokers.
R e su l t s R e su l t s D e m o g r a p h i c a n d Cl i n i c a l Ch a r a c t e r i s t i c s D e m o g r a p h i c a n d Cl i n i c a l Ch a r a c t e r i s t i c s As shown in figure 1, age distribution differs between outpatients and inpatients with younger patients in outpatients and older patients in inpatients.   T wo out of seven outpatients and 5/15 inpatients were heavy daily current smokers with a daily number of 20, or more, cigarettes (data unavailable for 1 inpatient). Current smoking status seems to confer a protective effect on the risk of contracting symptomatic COVID-19. However, 4.4% of inpatients and 5.4% of outpatients were daily current smokers, and among them two outpatients and five inpatients were heavy smokers (i.e. smoked 20 or more cigarettes a day), representing 1.5% of the cohort of COVID-19 patients in this study., Under or over-reporting of smoking status is a major concern of study on smoking habits. Actually, smoking status is more often reported in medical files in case of comorbidities. In our study, we have a very low rate of missing data on smoking status, and sensitivity analyses have shown that they do not alter the robustness of our results.
One pending question was "Does smoking prevent SARS-CoV-2 from infecting a person, or does it affect the severity of the disease?" Our study is the first one to include and analyze separately the smoking pattern of COVID-19 outpatients (non severe cases) and inpatients (severe cases). Indeed, all previous studies, but two, which reported smoking rates included only inpatients (T ABLE 1). Unfortunately, smoking data from inpatients and outpatients were mixed in the Guan study[1] and there was a lot of missing data in the report from the CDC [12]. It is of note that due to the low number of daily current smokers in our study, we could not conclude whether daily current smoking has an impact on the COVID-19 severity. However, SIR of inpatients did not differ from SIR in outpatient. In addition, the more severe COVID-19 patients, hospitalized in an ICU, were not included in the present study. A larger study including ICU patients will certainly help to conclusively address this question.
Our findings should be cautiously interpreted in light of its limitations. T he study was performed in 2020 and the results were compared to the data obtained from the French general population's smoking rate in 2018. However, we are not aware of such a dramatic trend of decrease in tobacco use in France since mid 2018 that could explain our results. T he SIRs were estimated with the assumption that the studied population who lives in a limited area around a Parisian hospital has the same smoking habits as the general French population. Actually, smoking rates differ across socio-professional categories, and therefore may differ across geographic areas. It is furthermore of note that in the present study healthcare workers are over-represented in the outpatient group, due to systematic testing at their work place for healthcare workers with symptoms, but not in the inpatient group (data not shown). It is, however, very unlikely that the very low SIRs that were estimated both for the out-and inpatient groups are the result of the study setting. Finally, in our study, smoking status was assessed only in symptomatic COVID-19 patients while a large part of infected individuals are asymptomatic [14] Because this is a cross-sectional study, we cannot confirm the causality of this association. We cannot also identify which of the many compounds of tobacco exerts the protective effect of smoking on COVID-19. T here are however, sufficient scientific data to suggest that smoking protection is likely to be mediated by nicotine. T he SARS-CoV2 is known to use the angiotensin converting enzyme 2 (ACE2) as a receptor for cell entry [15][16][17], and there is evidence that nicotine modulates ACE2 expression [18] which could in turn modulates the nicotinic acetyl choline receptor (manuscript submitted).
Inour hypothesis, the SARS-CoV2 would affect the control of the nicotine receptor by acetylcholine. T his hypothesis could also explain why previous studies have found an association between smoking and Covid severity[1, 9,10]. As hospitals generally impose smoking cessation and nicotine withdrawal at the point of hospitalization, tobacco (nicotine) cessation could release nicotine receptors, that are increased in smokers, and lead to a rebounding effect responsible for the worsening observed in hospitalized smokers.
T here could be significant implications of our results in terms of prevention, and for practice in hospitals. Nicotine administration, e.g. via a transcutaneous route might be a promising therapeutic option to recapitulate the protecting effect of smoking over SARS CoV2 infection in non-smokers, either as a prevention strategy, or as a curative strategy.
T he promotion of smoking to avoid COVID-19 is however not an option, this is obviously unacceptable, given the deleterious benefit/risk ratio of smoking which is responsible for a very heavy public health burden with more than 78 000 deaths per year in France.