## How much is a fifth?

Posted on Saturday 27 December 2008 at 7:38 pm by Jacob Aron
In

If a health risk is increased by a fifth, is that a lot? More importantly, should you worry about it? Many news outlets are reporting on a story from the World Cancer Research Fund that just one pint of beer a day, or other alcohol equivalent, can increase your risk of bowel cancer by a fifth. I’m going to pick on the Daily Mail in particular (mainly for the comments which I’ll get to later), but all of the stories are pretty much the same.

The question is, should I be worried enough to cut back on having a few pints, especially during this festive season? The important number here is not the relative risk (an increase of a fifth) but the absolute risk. According to Cancer Research UK, there are 61 diagnoses of bowel cancer in the UK for every 100,000 people each year. In other words, the chance of you getting bowel cancer is 61/100,000, or 0.061%.

Now, these statistics will include all instances of bowel cancer, including drinkers and non-drinkers alike, but for the moment let’s pretend that it’s only non-drinkers. Then, if everyone in the UK takes up drinking a pint a day, and thus risk of bowel cancer increase by a fifth for everyone, around 12 more people in every 100,000 will be diagnosed each year, corresponding to an absolute risk of 0.073%. I’m fudging the maths a little bit, because I don’t know how alcohol factors in to the Cancer Research UK data, but I’m actually making it look worse than it really is, because with accurate information on the effect of alcohol, the increase in risk would be even smaller. Remember, I’m making the (very wrong) assumption that no-one in the UK drinks!

In other words, when you look at the risk in absolute terms, it has hardly increased at all. Personally, those figures don’t worry me in the slightest. Yet, all of the mainstream media run the story with “beer makes you a fifth more likely to get cancer” because that is the eye-catching headline. The trouble is, we’re often giving conflicting information about whether drinking (in moderation) is “good” or “bad” for us, and this “flip-flopping” causes a cynicism of science apparent in both the Daily Mail’s headline (“Cheers! Now they tell us beer and wine give us cancer”) and it’s commenters. A typical example is this comment from Bryan Caffyn:

Please when will they,the so called experts, make up their minds, last week we were being told by the very same people, a glass or two of wine would reduce the risk of all sorts, now its going to increase the chance of bowel or liver cancer. Even by this barmy bunches standards this is crazy, time gormless gordon dare I say took the lead, no I musnĀ“t be silly, Christmas is over.

What these stories don’t get across is that most substances we consume are both improve and are detrimental to our health. The science isn’t wrong (I’m assuming, of course, having not read any actual papers); there really is an increased risk of bowel cancer from drinking. That doesn’t mean that drinking can’t also have beneficial effects in other ways. When the changes in absolute risk are so small, however, who really cares? A fifth of relative risk just isn’t enough to be worth worrying about!

Sorry, comments for this entry are closed at this time.