Sir,

The ongoing outbreak of coronavirus disease 2019 has become a challenge for researchers and health systems.¹ Recently, Cai commented on Lancet Respiratory Medicine that COVID-19 prevalence is dependent, among other variables, on sex and smoking.² Brake³ demonstrated that smoking upregulates the expression of Angiotensin-Converting Enzyme-2 (ACE2) receptor in type-2 pneumocytes of smokers with COPD, increasing the ability of this coronavirus to infect the human cells. The e-cigarettes and heat-not-burn (HNB) devices (e.g.& JUUL) also have been involved with the ACE2 overexpression and higher prevalence of viral respiratory infections (e.g.& influenza).

Unbelievably, the smoking status is uncommonly reported in COVID-19 cases hampering our ability to establish any association between this potential risk factor with the disease. A rapid review of PubMed using search terms as smoke, epidemiology, prevalence, COVID-19, coronavirus, 2019-nCoV retrieved zero articles to date. Whether& smoking rates are related to the COVID-19 remains unknown. Thus, we searched into the World Bank Open Data (https://data.worldbank.org/indicator/SH.PRV.SMOK) and the WHO global report on tobacco smoking⁴ that the total population per country, number of people >65 years old, prevalence of smoking (including new devices) of the top 25 countries on the COVID-19 list.

A Pearson’s correlation matrix was performed (Table 1), a strong uphill correlation was observed between the total number of COVID-19 with the total population in each country (r=0×68), the population living with >65 years old (r=0×74), and the number of smokers (r=0×69). After this exploratory analysis, a linear regression model using OLS was fitted to the dataset (Graphic 1). An empirical linear relationship (R²= 0×97) was observed between the population older than 65 years old, the number of smokers and the total number of confirmed COVID cases in the top 25 countries.

Therefore, a lower progression of COVID-19 could be expected in demographics with lower tobacco use (e.g.& Colombia), if isolation, quarantine and other measures are extensively used¹. Obviously, other confounding factors may explain this relationship, but this hypothesis may be a starting point for exploration in current surveillance databases and to clarify the true magnitude and direction of this association.

We urgently need more data, while the healthcare authorities should increase the surveillance of the smoking status in diagnosed COVID-19 patients, and promoting stopping tobacco use during the pandemic. Notably, in Chinese patients diagnosed with COVID-19, the odds of disease progression and death were 14 times higher among smokers compared to non-smokers.⁵