Wednesday 7 December 2011

Cancer and lifestyle choices

The headlines are startling: over 40% of cancers are down to lifestyle. The opening paragraph of the BBC News report states it bluntly:
Nearly half of cancers diagnosed in the UK each year - over 130,000 in total - are caused by avoidable life choices including smoking, drinking and eating the wrong things, a review reveals.
For those with cancer, or who have family or friends suffering from the disease, this makes for grim reading. Because, like it or not, there's a moralistic sub-text here: in over 40% of cases, cancer is the fault of the patient. If only they'd chosen the right life-style, they wouldn't be sick.

Now, that isn't what the articles say, but it's a clear implication. Before we go any further though, perhaps we ought to back up a little and ask what the science really says...

Firstly, we have to be clear that what we're talking about here are projections based on a series of mathematical models. These projections take a series of risk factors - say smoking or alcohol consumption - and then compare the increased risks of developing cancer based on comparing current consumption levels and what are considered the 'optimum' levels. So, for example, you take the additional risk of developing cancer per unit of alcohol, the difference between current consumption and the ideal level and calculate the increased risk. Now take that increased risk and multiply by population and you get your projected number of cancer patients. It's a bit more complicated than that, of course, but in principle that's how it works. Add an additional set of factors for different types of cancer, and you've got a cascade of different variables that make up that final scary number.

So, all of this is subject to many different assumptions about risk factors, current consumption levels, ideal consumption levels, population growth etc.
This is not the same as saying that we can pinpoint which cancers were caused by life-style factors. We can't take 100 cancer patients and say these 40 have the disease because they smoked, drank, were fat or didn't eat enough fruit and veg. It doesn't mean that we can point to a person and say that if they had had a better diet or were tee-total then they wouldn't have cancer, or even that there'd be a 60% lower chance of them having cancer.

Even with smoking and cancer - which is the biggest risk factor by far, and the one for which we have the most evidence, both at an epidemiological and biological level - there are people who smoke for decades and don't get lung cancer, just as there are people who have never smoked and get the disease anyway. But let's be clear, smoking is not a good idea, not under any circumstances. For some of the other risk factors, such as obesity, fruit and vegetable consumption, drinking alcohol and so on, we have much less confidence in those numbers. And remember, these other risk factors come a long way after the risks associated with smoking - a very long way.

So, what can we conclude from these projections? Do they give us any useful information? Yes. Even if we can't pin-point which individuals have got cancer because they didn't get enough fruit and veg, it does suggest that in general a healthier diet should reduce the number of people in the population who get cancer. And it underlines yet again the dangers of smoking. So, when talking about the population as a whole, the projections can be useful indicators, but when we are dealing with individuals they don't necessarily tell us much. Always keep in mind that in these types of studies we are talking about population models, not what we actually know about people who have got cancer at the moment.

Indeed, in the small print the language of the researchers is a lot more cautious than in the headlines. For example, it is worth pointing to the caveats listed on the Cancer Research UK web site:
Quantifying the likely impact of preventive interventions requires complex scenario modelling. So for example, while 50% of colorectal cancer cases are attributable to lifestyle (diet, alcohol, physical inactivity and overweight), it has been estimated that only about half of this number are preventable over a 20-year period.
In plain English this means that even if those patients with colorectal cancer who are likely to have been adversely affected by diet, drinking etc had switched to a better life-style, 50% of them would still have got cancer. You see, it's just not as straightforward as the news headlines like to suggest.

Unfortunately these subtle nuances get lost in the headlines. And although nobody comes out and says that 40% of cancer patients have got nobody to blame but themselves, it's there between the lines. Aside from making cancer patients feel guilty, we can also see that there is a political dimension in all of this. It's this sort of study that is used as the 'evidence base' for driving through public health policy decisions. This study will be used to justify initiatives on diet, exercise, health-awareness campaigns etc. The projections will be taken as fact and will be used to make laws that are intended to correct our incorrect lifestyles. The fact that this is a model will be ignored when it comes to the political discussions on public health.

And finally, let's not lose sight of the other 60%. If 40% of cancers are associated with life-style factors, that still leaves 60% where there is no such association.

The bottom line is that cancer is a complex disease with many factors that contribute to it. A healthy diet, not smoking, drinking in moderation and doing regular exercise are all good, and may contribute to a lower chance of getting cancer. We know that if the population ate more fruit and veg, stopped smoking, did more exercise etc then the number of people getting cancer would probably decrease. That's it, that's all we can say. It does not even mean that you can reduce your individual chance of getting cancer by 40% by following a better life-style. And it certainly does not mean that 40% of cancer patients are to blame for their disease.

UPDATE: As I suspected, there are many cancer patients complaining that they are being blamed for their disease. It's there in the comments on the Cancer Research UK blog, on the Guardian, Daily Telegraph, BBC and other sites. The reporting of this story has been pretty appalling really.

No comments:

Post a Comment