Smoking and COVID-19 - what's new?

Forgive me for boasting, but I reached a personal milestone last month when an article of mine was classified as fake news by Facebook. The article discussed some of the many studies that have emerged in the last two months showing smokers to be significantly under-represented in the coronavirus wards of Europe, America and China. Those who stumble across on it on Facebook are now warned that it is ‘misleading’.

Admittedly, the headline was ‘Smoke fags, save lives’ and I did call on doctors to start prescribing Lucky Strikes. Perhaps I overstepped the line when I called for the British public to clap for cigarettes every Thursday evening. But surely the intelligent reader could tell these comments were tongue-in-cheek?

Joking aside, the hypothesis that smokers are at less risk of contracting COVID-19 has a growing weight of evidence behind it and is being taken seriously by serious people. There are now more than twenty studies pointing in that direction. In the last week, two meta-analyses combining findings from the scientific literature have been published. One of them suggested that smokers are 78 per cent less likely to ‘have an adverse outcome’ from COVID-19. The other suggested that smokers are 82 per cent less likely to be hospitalised with the disease in the first place. 

This is an enormous protective effect, if true. But is it? As you might expect, the medical establishment is not exactly punching the air with joy at this news and every alternative explanation is being thoroughly explored. There are certainly reasons to be sceptical. Some of the studies have not yet been peer-reviewed and most of them rely on crude comparisons between the number of smokers with COVID-19 and the number of smokers in the general population. A large study from the UK last week found that only 6.9 per cent of the people who died from the virus were smokers, but this translated into a slight reduction in risk, a slight increase in risk or no significant effect at all, depending on how the figures were adjusted for other factors.

It is also likely that the unfashionable habit of smoking is under-reported by people attending hospital with an acute respiratory disease, and yet it is difficult to believe that ‘shy smokers’ fully explain the exceptionally low rates of smoking recorded among COVID-19 cases. The percentage of smokers with the virus is typically around five per cent in countries where the smoking rate is between 15 and 30 per cent. Since COVID-19 attacks the lungs and preys on those with underlying health conditions, one would expect smokers to be over-represented in these studies. Instead, they are heavily under-represented.

It is natural to approach counter-intuitive findings with caution, but there is a difference between scepticism and blind denial. When asked about the smoking-coronavirus link on television recently, the celebrity doctor Xand van Tulleken said: ‘I haven't looked into this particular piece of research but I would discount it completely. It is definitely wrong.’ Public Health England’s only contribution has been to assert that smokers who contract COVID-19 are fourteen times more likely to die from it, based on a tiny Chinese study which included just five smokers. It is notable that those who are so eager to point out the potential flaws in the epidemiological evidence on smoking and coronavirus happily cite exactly the same kind of evidence when talking about obesity and coronavirus.

In any case, not all the studies have relied on a crude comparison with the national smoking rate. Studies from the USA and France have adjusted for other factors - including, crucially, age - and found that the significant protective effect still holds.

The working hypothesis is that nicotine is responsible. This is assumed because there are plausible biological mechanisms by which nicotine could confound the virus and because the prospect of smoking per se being the prophylactic is too terrible for doctors to contemplate. But it remains only an assumption. The French neuroscientist Jean-Pierre Changeux is currently carrying out experiments with nicotine patches, but he has yet to publish his findings. Other than that, the only real crumb of evidence comes from a study based on an online survey which concluded that vapers were 55 per cent less likely to get COVID-19.

Although smokers are less likely to be hospitalised with COVID-19, several studies have found them to have worse outcomes once admitted. With smoking banned in and around hospitals, this leaves open the intriguing possibility that smokers are less able to fight the virus once they are deprived of nicotine. This idea was raised by doctors at the Royal Glamorgan Hospital back in mid-March, before the lockdown began. They suggested that ‘the simple use of nicotine patches should be be urgently considered and discussed’. This advice fell on deaf ears, but perhaps giving nicotine patches to COVID-19 patients is not such a bad idea - and not just to smokers.


Previously published in the Telegraph

I have updated my list of smoking/coronavirus studies here.