In the late 1940s, Austin Bradford Hill and Richard Doll began interviewing hundreds of hospital patients, half of whom had lung cancer while the other half had various other diseases. Their research, published in 1950, showed that 99.7 per cent of the male lung cancer patients had a history of smoking. Out of 649 patients, only two were nonsmokers. In retrospect, it seems amazing that no one had spotted the link between smoking and lung cancer before, but it is less surprising when you consider that 94.8 per cent of the men who were not in hospital with lung cancer also had a history of smoking.
This might not seem like such a big difference. The vast majority of the men in hospital had smoked tobacco, regardless of what they were being treated for. And yet a statistician can tell from the figures above that smokers were fourteen times more likely to get lung cancer than nonsmokers.
Richard Doll continued his research into smoking for the rest of his life. In 2004, he published a study which followed a group of doctors for fifty years and found that those who smoked were fifteen times more likely to die from lung cancer than the nonsmokers, a remarkably similar result to that found in his first study.
This is worth remembering when you see figures showing the number of vaccinated and unvaccinated people in hospital with Covid-19. A report from Public Health England shows that 84 per cent of patients in hospital with Covid aged 70 or over had been double-jabbed whereas “only” 13 per cent of them were unvaccinated. Anti-vaxxers leap on such statistics as proof that the vaccines don’t work, but these figures mean nothing unless we know the denominator.
Among people aged 70 or over, 94 per cent have been double-jabbed. The difference between 84 per cent and 94 per cent might not seem that great, but it means that unvaccinated elderly people are three times more likely to be hospitalised with Covid-19 than those who are vaccinated. Getting vaccinated reduces their chance of being hospitalised by two-thirds. At younger ages, the reduction is greater still.
But the most important denominator is the number of people being infected. There have been more infections since June than there were between December and March. In those dark winter months, there were 58,000 Covid-related deaths in England. In the last four months, despite more than two and a half million infections, there have been 7,000. This is the vaccine at work.
As we learn to live with an endemic virus, there is one more denominator to bear in mind: the number of people dying every week. In recent weeks, it has hovered at around 100 a day in England. What we don’t hear so much about is the number of non-Covid deaths, which is typically in the region of 1,300 a day.
Every week, 10,000 people die in England. More than half a million people die every year. It is difficult to visualise so many people. However, it is quite easy to imagine 100 people. For human beings who evolved to be around small groups, it seems like a big number, but in the context of the number of people who die each day, it really isn’t.
People tend to overestimate the risks of air travel and underestimate the risks of road travel because of “availability bias”. Plane crashes make headlines. Car crashes usually do not. It made sense to focus on the number of Covid deaths when the pandemic was in full swing, but unless they rise alarmingly in the coming weeks we need to start weaning ourselves off daily death counts.