Smoking increases risk of a poorer outcome in patients with COVID-19
By Nicole MacKee
Smokers have a higher risk of severe COVID-19 outcomes, including hospitalisation and death, UK researchers have reported in Thorax.
A large observational study of more than 420,000 participants – teamed with a Mendelian randomisation analysis of more than 280,000 participants to reduce confounding – found that when compared with never-smokers, current smokers had a higher risk of hospitalisation from COVID-19 (odds ratio[OR], 1.80) and at least a twofold increased risk of mortality (OR, 2.14 among mild smokers), which increased further with greater number of cigarettes smoked per day.
Earlier studies had suggested a lower prevalence of COVID-19 infection and death among smokers, the authors of an accompanying editorial noted.
Professor Christine Jenkins, Head of the Respiratory Group at the George Institute for Global Health, Sydney, said the study added to the evolving understanding about the association between smoking and COVID-19 outcomes, but the findings should be interpreted with caution.
‘A lot of things about COVID-19 are not what we expect,’ Professor Jenkins said, noting that original suppositions about risk factors were based on the knowledge of other major respiratory viruses, such as influenza virus, SARS-CoV-1 and Middle East respiratory syndrome coronavirus (MERS-CoV). ‘We have had those suppositions challenged.’
Professor Jenkins said although this study provided more certainty that smokers had an increased risk of more severe COVID-19 outcomes, the degree of risk remained unclear.
‘How much more likely are you to have a severe outcome if you’re a smoker – independent of whether you have cardiovascular disease, hypertension or diabetes, or are aged 75 – I don’t know, and I don’t think this study helps us, even though statistical adjustments were made to account for some of these risk factors.’
Although the Thorax researchers reported an up to sixfold increased risk of death among heavy smokers (20 or more cigarettes daily) compared with never-smokers, Professor Jenkins pointed to several limitations in the study that reduced confidence in this finding.
First, she said, the researchers had identified a relatively low COVID19 infection rate – 1649 infections in more than 421,000 patients. ‘That is a 0.4% infection rate, which seems exceptionally low for the UK where there was rampant COVID,’ she said, noting that this suggested a low community testing rate.
Also, Professor Jenkins said the findings may not be generalisable because the smoking rate among study participants was 3.3%, whereas the rate among the general UK population was 14%.
Professor Jenkins said, however, that the overall findings were sufficiently robust to give clinicians confidence to advise patients that smoking increased their likelihood of a poorer outcome should they contract COVID.
‘It’s one of the levers you can try to pull to help people not to smoke,’ she said. ‘Vaccination is by far the most important measure but smoking cessation, or not taking up smoking, is a really valuable addition.’
Also, Professor Jenkins said, smoking was an important red flag in triaging patients admitted to hospital with COVID. ‘It adds to the risk assessment for the patient who may not do well.’
Thorax 2021; doi:10.1136/thoraxjnl-2021-217080.
Thorax 2021; doi:10.1136/thoraxjnl-2021-217685.