A Canadian study available on the preprint server medRxiv* shows that current smokers are slightly more likely than never smokers to test positive for coronavirus disease (COVID-19), but cautions about the possibility of reverse causality where smokers actually quit to avoid the disease.
Current COVID-19 pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), may result in a life-threatening respiratory condition in a significant portion of infected individuals, primarily elderly and those with comorbidities.
Smoker to non-smoker ratio in individuals hospitalized early during the pandemic in China, the United States, and France varied. However, habitual smokers seemed to be markedly underrepresented among the group necessitating hospital treatment.
Some French researchers even suggested the so-called ‘nicotinic hypothesis’ for COVID-19, highlighting current smoking status as a protective factor with further preventative and therapeutic implications.
As the proper assessment of smoking risks in correlation with COVID-19 is best done in prospective studies, Dr. Eo Rin Cho and Dr. Arthur S. Slutsky from Unity Health Toronto hospital network, together with Dr. Prabhat Jha from the University of Toronto, decided to analyze obesity, smoking and other risk factors in the UK Biobank study.
Biobank data under the magnifying glass
The UK Biobank project is a prospective study of 502,506 adults aged between 40 and 86, followed for an average period of 11.3 years. It provides in-depth health, physical, and genetic data of individuals all across the UK.
Open access is a vital ingredient of this database, as scientists around the world can delve into the UK Biobank data directly and swiftly build on the work of other scientists – resulting in unanticipated dividends in diverse research fields.
And now it helps enormously in COVID-19 research. Between March 16 and April 26, 2020, a total of 2,237 participants were tested for SARS-CoV-2 infection; 40.6% of those tested positive, which means testing was done on patients at high risk of disease.
Furthermore, the smoking prevalence at enrollment was slightly lower in those with a positive SARS-CoV-2 test result (16.6%) in comparison to those testing negative (17.5%). However, these percentages reveal no relevant information before further statistical scrutiny.
Controlling for variables
Consequently, the Canadian authors conducted logistic regression analyses (a type of predictive analysis) controlling for age at the time of COVID-19 testing, as well as the following variables: smoking, body mass index (BMI), and self-reported hypertension or diabetes. Those who developed respiratory, vascular, or malignant diseases were excluded.
In sex-stratified analyses, current smokers had higher odds of being tested; also, former smokers, those with a history of hypertension or (among men) history of diabetes, and those with BMI over 27 were also more likely to be tested.
After adjusting for these variables as well, current smokers were slightly more likely than those who never smoked to test positive COVID-19. Still, this was not statistically significant, and former smoking was similarly not associated with a positive test.
The need for more data
“Further examination of smoking as a risk factor for COVID-19 in other prospective studies is required”, caution study authors in the paper available on medRxiv preprint server.
“This must take into account not only reverse causality, where smokers quit to avoid disease, but also prior diseases and comorbidities including obesity, diabetes, and hypertension, as each is associated with COVID-19 hospitalization or mortality”, they add.
As of May 2020, the COVID-19 death toll has thus far risen to almost 300 thousand, which reflects approximately 25 days of annual global tobacco deaths. Therefore, both of these lethal epidemics should be taken seriously, and their burden measured appropriately.
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.