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

Background: SARS-CoV-2 is the causative agent of COVID-19, an emergent zoonotic disease which has reached pandemic levels and is designated a public health emergency of international concern. It is plausible that former or current smoking status are risk factors for infection, hospitalisation and/or mortality from COVID-19. Objective: We aimed to estimate the rates of i) infection, ii) hospitalisation, iii) disease severity, and iv) mortality from SARS-CoV-2/COVID-19 stratified by smoking status. Methods: We adopted recommended practice for rapid evidence reviews, which involved limiting the search to main databases and having one reviewer extract data and another verify. Published articles and pre-prints were identified via MEDLINE, EPPI-Mapper and expertise within the review team. We included observational studies with community-dwelling or hospitalised adults aged 16+ years who had been tested for SARS-CoV-2 or were diagnosed with COVID-19, providing that data on smoking status were reported. Studies were judged as high quality if they explicitly recorded current, former and never smoking status with low levels of missing data. Results: Twenty-eight studies were included, 22 of which were conducted in China, three in the US, one in Korea, one in France and one across multiple international sites with data predominantly collected in the UK. Eight studies did not state the source for information on smoking status. Twenty-five studies reported current and/or former smoking status but had high levels of missing data and/or did not explicitly state whether the remaining participants were never smokers. Notwithstanding these uncertainties, compared with national prevalence estimates, recorded current and former smoking rates in the included studies were generally lower than expected. high-quality a fixed-effects meta-analysis no evidence for an increased risk of hospitalisation among 657 current/former smokers tested positive in the community (RR = 1.03, 95% CI = 0.93-1.14, p = 0.57). 1370 people hospitalised across two high-quality studies, greater disease severity in current/former smokers compared with never smokers (RR 1.43, 95% CI 1.15-1.77, p = .002). on never smoking


Introduction Introduction
COVID-19 is a respiratory disease caused by the emerging SARS-CoV-2 virus. Large age and gender differences in case severity and mortality have been observed in the ongoing COVID-19 pandemic 1 ; however, these differences are currently unexplained. SARS-CoV-2 enters epithelial cells through the ACE-2 receptor 2 . Some evidence suggests that gene expression and subsequent receptor levels are elevated in the airway and oral epithelium of current smokers 3,4 , thus putting smokers at higher risk of contracting SARS-CoV-2.
Other studies, however, suggest that nicotine downregulates the ACE-2 receptor 5 .
T hese uncertainties notwithstanding, both former and current smoking is known to increase the risk of respiratory viral 6,7 and bacterial 8,9 infections and is associated with worse outcomes once infected. Cigarette smoke reduces the respiratory immune defence through peri-bronchiolar inflammation and fibrosis, impaired mucociliary clearance and disruption of the respiratory epithelium 10 . T here is also reason to believe that behavioural factors (e.g. regular hand-to-mouth movements) involved in smoking may increase SARS-CoV-2 infection and transmission in current smokers. However, early data from the COVID-19 pandemic have not provided clear evidence for a negative impact of current or former smoking on SARS-CoV-2 infection or COVID-19 disease outcomes, such as hospitalisation or mortality 11 . It has also been hypothesised that nicotine might protect against a hyper-inflammatory response (or "cytokine storm") to SARS-CoV-2 infection, which may lead to adverse outcomes in patients with COVID-19 disease 12.
T here are several reviews that fall within the scope of smoking and COVID-19 11,13-17 .
We aimed to produce a rapid synthesis of available evidence pertaining to the rates of infection, hospitalisation, disease severity and mortality from SARS-CoV-2/COVID-19 stratified by smoking status. Given the increasing availability of data on this topic, this is a living review with regular updates. As evidence accumulates, the review will be expanded to include studies reporting COVID-19 outcomes by alternative nicotine use (e.g., nicotine replacement therapy or e-cigarettes). the total number of infections in a given region/city is unknown. With a largely unknown disease process, it is also difficult to determine whether the time between the exposure and outcome is sufficient. We therefore focused on three of the 14 criteria to determine whether studies were of sufficient quality to warrant inclusion in meta-analysis. Studies were judged as 'good' quality if they: i) had low levels of missing data on smoking status (i.e. <20%) and used a reliable self-report measure that distinguished between current, former and never smoking status, ii) used biochemical verification of smoking status, and iii) adjusted smoking and COVID-19 analyses for covariates that are likely to confound these associations (e.g. age, non-smoking related comorbidities). Studies were rated as 'fair' if they fulfilled criterion i) and were otherwise rated as 'poor'. T he quality appraisal was conducted by one reviewer and verified by a second.

Evidence synthesis
A narrative synthesis was conducted. Where possible, data were pooled in R v.3.6.3 21 with the Mantel-Haenszel or inverse variance method using random or fixed effects, depending on heterogeneity, and presented as risk ratios (RRs) 22 . Heterogeneity between study outcomes was assessed using the I 2 statistic, suitable for smaller metaanalyses 23 .
T o aid in the visualisation of smoking prevalence in the included studies, 95% bootstrap percentile confidence intervals were calculated. We performed 1,000 bootstrap replications, with the 2.5th and 97.5th percentiles of the empirical distribution forming the 95% bootstrap percentile confidence intervals 24 . Prevalence estimates in the included studies were not adjusted for age, sex or socioeconomic position.

Results Results
In the current review version (v4), a total of 456 new records were identified, with 102 studies included in a narrative synthesis and 16 studies included in meta-analyses (see Figure 1). Use of alternative nicotine products T wo studies recorded the use of alternative nicotine products in current and/or former smokers but did not report COVID-19 outcomes stratified by nicotine use.

Quality appraisal
Nineteen studies were rated as 'fair' quality due to having low levels of missing data and distinguishing between current, former and never smoking status (see Appendix 1). T he remaining 83 studies were rated as 'poor' quality.

Smoking prevalence by country
Unadjusted smoking prevalence in community and hospital studies compared with overall estimates for national adult smoking prevalence split by country is presented in Figure 2a and 2b. Lower than expected current and former smoking prevalence was generally observed, with similar to expected prevalence reported in studies conducted in the UK.
Variability in prevalence estimates was observed across studies conducted in the US.

Discussion Discussion
T his rapid review of 102 studies found substantial uncertainty arising from the recording of smoking status. Notwithstanding these uncertainties, compared with overall adult national prevalence estimates, recorded current and former smoking rates in most countries were lower than expected. From available data, there was insufficient evidence to conclude that current and/or former smoking status is associated with hospitalisation or mortality. T here was some evidence from 'fair' quality studies that the risk of SARS-CoV-2 infection is reduced, and disease severity in those hospitalised for COVID-19 is greater, in current but not former smokers compared with never smokers.

Infection by smoking status
In seven 'fair' quality studies, evidence suggest that current (but not former) smokers in the community are less likely to test positive for SARS-CoV-2 compared with never smokers. It should be noted that criteria for accessing testing will vary during the course of the epidemic. It is possible that current and former smokers are more likely to receive a test due to increased prevalence of cough or altered sense of smell or taste 125 , which are used as screening criteria. Infection positivity rates estimated among random Qeios, CC-BY 4.0 · Article, June 11, 2020 Qeios ID: UJR2AW.5 · https://doi.org/10.32388/UJR2AW.5 19/35 samples will be more informative than currently available data. Smoking status is being collected in at least one large representative infection and antibody survey in the UK 126 . Hospitalisation and disease severity by smoking status As reported elsewhere16, smoking prevalence among multiple hospital and community cohorts was consistently lower than national estimates from China, USA, France, Spain, Israel, Italy, Mexico, Finland, Iran, Kuwait and Switzerland. In a single study conducted in Korea and nine studies of varying quality conducted in the UK, however, current and former smoking rates were similar to those expected.
In five 'fair' quality studies across four countries, there was no evidence that current or former smokers are at lower risk of hospitalisation for COVID-19 compared with never smokers among those identified as testing positive. T here was some evidence from four 'fair' quality studies that current smokers are at increased risk of greater disease severity compared with never smokers.

Mortality by smoking status
In three 'fair' quality studies, there was inconsistent evidence on the association of smoking status and the risk of death from COVID-19. Given lack of knowledge of the disease progression and long-term outcomes of COVID-19 disease, it is unclear whether studies conducted thus far in the pandemic have monitored patients for a sufficient time period to report complete survival outcomes or whether this reflects early censoring.

Issues complicating interpretation
Interpretation of these early studies is complicated by several factors (see Figure 11).
First, exposure to SARS-CoV-2 is heterogeneous with different subgroups being at heightened risk of infection at different stages of the pandemic. T his will likely introduce bias in studies assessing the rate of infection by smoking status conducted early on.
Second, as mentioned above, current and former smokers may be more likely to meet local criteria for community testing due to increased prevalence of symptoms consistent with SARS-CoV-2 infection, such as cough, increased sputum production or altered sense of smell or taste125. T hird, testing for acute infection requires swabbing of the mucosal epithelium, which may be disrupted in current smokers, potentially altering the sensitivity of assays 87 .  Limitations T his rapid review was limited by not having two independent reviewers extracting data, limiting the search to one electronic database and one pre-print server and by not including at least three large population surveys due to their reliance on self-reported SARS-CoV-2 infection (which means they are not meeting our eligibility criteria) 125,131,132 .
Population surveys -particularly with linked health data -will be included in future review versions to help mitigate some of the limitations of healthcare based observational studies. T he comparisons with national estimates of smoking prevalence did not adjust observed prevalence for the demographic profile of those admitted/tested. Other reviews focused on this comparison have applied adjustment for sex, and continue to find lower than expected prevalence -notwithstanding the issues complicating interpretation described above 16 .

Implications for research, policy and practice
Further scientific research is needed to resolve the mixed findings summarised in our review. First, clinical trials of the posited therapeutic effect of nicotine could have important implications both for smokers and for improved understanding of the SARS-CoV-2 virus. Such trials should focus on medicinal nicotine (as smoked tobacco is a dirty delivery mechanism that could mask beneficial effects) and potentially differentiate between different modes of delivery (i.e. inhaled vs. not) since this can affect pharmacokinetics 133 and potential therapeutic effects. A second research priority would be a large, representative (randomly sampled) population survey with a validated assessment of smoking status which distinguishes between recent and long-term exsmokers -ideally biochemically verified -and assesses seroprevalence and links to health records.
In the meantime, public-facing messages about the possible protective effect of smoking or nicotine are premature. In our view, until there is further research, the quality of the evidence does not justify the huge risk associated with a message likely to reach millions of people that a lethal activity, such as smoking, may protect against COVID-19.
It continues to be appropriate to recommend smoking cessation and emphasise the role

Conclusion
Across 102 studies, there is substantial uncertainty arising from the recording of smoking status on whether current and/or former smoking status is associated with SARS-CoV-2 infection, hospitalisation or mortality. T here is some evidence that current smoking compared with never is associated with reduced risk of testing positive in the community but greater disease severity in those hospitalised for COVID-19.