Do people in “public health” have little meetings where they dare each other to tell the media the most outrageous nonsense they can think of? Is it a competition? Do they put money in a pot which they only lose if a journalist laughs at their press release and refuses to publish it?

If so, that day never seems to come. I stopped giving money to Cancer Research UK (CRUK) years ago when it became obvious that they were prepared to abuse people’s trust in their brand by using dodgy claims to lobby for stupid policies. I suppose you have to expect activists to gild the lily somewhat, but there’s gilding and there’s gaslighting.

Rishi Sunak’s plan to very gradually prohibit the sale of tobacco had to be put on hold when he pulled off the political masterstroke of holding a general election in July for no reason. It is extremely unlikely that the Labour Party will not revive it — they were keener on the idea than the Tories — but the anti-smoking lobby are not taking any chances and are lobbying hard for it to be made a priority. The only snag is that a mere six per cent of the general public share their belief that it is a priority. And so, to inject some urgency into the proceedings, CRUK announced on Tuesday that “the number of cancer cases caused by smoking in the UK has reached an all-time high”. 

Does anybody in their right mind find this remotely believable? Smoking rates in the UK peaked at around 60 per cent in the 1950s. By 1990, they had halved to 30 per cent. They have since more than halved again, to 13 per cent. It takes a while to develop cancer, of course, but it doesn’t take 70 years. If it did, you might as well smoke. 

When the renowned epidemiologists Richard Doll and Austin Bradford Hill looked at this in 2000, they found that rates of smoking and lung cancer mortality rose and fell among both men and women with a time lag of 20-30 years. Indisputable evidence from around the world backs this up. Lung cancer is not the only type of cancer that is caused by smoking, but it is by far the most common. According to CRUK, it accounts for 58 per cent of smoking-related cancers. The others also have a time-lag, but it is typically in the same ballpark as the 20-30 years it takes lung cancer to develop. There is simply no way that the steep and steady decline in the smoking rate that has been going on since the early days of Queen Elizabeth II’s reign has not fed through into lower rates of smoking-related cancer by now.

CRUK are such a trusted source of information that they did not feel the need to show their workings

Promoting this ludicrous factoid on the Today programme was a woman from CRUK who claimed that the counterintuitive rise in smoking-related cancers can be explained by the fact that the population has grown. There are, she said, still 6.4 million smokers in the UK. This is perfectly true, but she neglected to mention that there were 10 million of us puffing away in the 1990s and nearly 20 million in the early 1970s. Her other explanation was “lag time, actually” which she elaborated on by saying that “the cancers caused by smoking today actually reflect what the smoking levels were like ten, fifteen years ago rather than what they are now”. The presenter Nick Robinson missed the open goal of pointing out that smoking rates were lower fifteen years ago than they were thirty years ago and much lower than they had been sixty years ago, and that all of this did not seem consistent with people in 2024 dropping dead from smoking-related diseases at an unprecedented rate. Instead the conversation turned to the all-important question of whether ethnic minorities and members of the LGBTQ+ community are at greater risk of smoking tobacco. 

CRUK are such a trusted source of information that they did not feel the need to show their workings. The underlying research is, they say, “in preparation”. If I assumed good faith and had to guess, I would say that they have made a terrible mess of their attributable fractions. Attributable fractions are how we estimate how many cases of a disease are caused by a particular risk factor. Since doctors do not keep a record of how many of their patients get ill as a result of various lifestyle factors (and they wouldn’t know for sure anyway), we take the total number of cases and assume that a certain proportion of them are smoking-related, alcohol-related, obesity-related and so on. This is not an exact science, to say the least, and it is highly vulnerable to changing circumstances and changing assumptions. In 2021, the number of alcohol-related deaths plummeted by 23 per cent as a result of the attributable fraction system being updated. People who think that deaths from COVID-19 were defined in a questionable manner would have a stroke if they understood how deaths from smoking and obesity are counted.

According to CRUK, there are 16 types of cancer that are “smoking-related” (they added breast cancer to the list on Tuesday for good measure). CRUK used to say that one in three of us would get cancer at some point in our lives. These days, they say the figure is one in two. It is one of the paradoxes of public health that as people lead healthier lives they become more likely to get cancer. That is because age is the biggest risk factor for most cancers. 

The incidence of cancer rose by 41 per cent between 2002 and 2019, according to Macmillan Cancer Support (a charity that is more deserving of your donations). As the population ages, there will be more and more people getting the disease. And, unless CRUK sorts its methodology out, an implausibly large number of them will be getting it as a result of smoking, even though they have never smoked.