For regular readers, alcohol and breast cancer may seem like old news, and you might wonder why it’s hit the headlines again.
But solid new data from a team of international researchers gives a new, more accurate estimate of how much one small drink a day can increase a woman’s risk of breast cancer.
The study – which reviews all the available evidence – helps settle the question of how big the risk is for women who only drink lightly, as we’ll see below.
The research also increases the overall strength of the evidence for a link between alcohol and breast cancer.
And given the recent news on minimum alcohol pricing, it’s worth bearing in mind the impact that small changes can make.
Why is this new?
Research has already established beyond doubt that breast cancer is more common in women who regularly drink alcohol. And we also know that the more alcohol a woman drinks, the greater her chances of developing the disease.
But there’s still been a question about the size of the risk for women drinking smaller amounts of alcohol – and different studies have given slightly different answers.
Until now, the best evidence available for the risk in light drinkers has come from doing ‘dose-risk analyses’.
In these, researchers plot all the results from a study on a graph – in this case for women drinking a certain amount of alcohol (dose), what proportion developed breast cancer (risk) – and find an equation that best describes how the two things are related.
The results from this type of analysis have shown an increase in risk of breast cancer of around 7 – 12 per cent for each extra drink a day.
But this isn’t the same as actually measuring that change in risk in women who drink lightly.
And because the change in risk is quite small, studies that set out to directly measure this risk couldn’t be sure their results weren’t down to chance, whether they showed an increase in risk or not.
This new study is what’s known as a meta-analysis. It looked at all the available results of research that directly measured the effect of light drinking on breast cancer – a total of 113 academic papers – and used statistical analysis to reveal the pattern behind all the data.
It showed that the risk of breast cancer in women drinking one small drink (about 1 ½ units) a day increased by 5 per cent. This figure was widely reported by the media.
But what does this figure mean? 5 per cent of what? To really understand what the research is saying, we need to turn this into an ‘absolute risk’ figure.
Here comes a tiny bit of maths
To convert a relative risk into an absolute risk figure, we need to know the overall risk of women who don’t drink (since this is the figure that increases by 5 per cent).
Unfortunately, this paper doesn’t contain that statistic. However, we can make an educated estimate:
As we blogged about last year, the lifetime risk of breast cancer for all women is 1 in 8, or 12.5 per cent. This means that 12-and-a-half women in every hundred (or, better, 125 out of 1000, since you can’t have half a woman) develop breast cancer at some stage in their lives.
But this figure includes all women, from non-drinkers to those who drink very heavily. So let’s make an assumption* that the lifetime risk of non-drinkers is lower – 1 in 9, or 11.1 per cent.
If we multiply this figure by 5 per cent (the increase due to low level drinking), we get 11.7 per cent.
Now let’s convert these numbers to actual people
In a thousand non-drinkers, we can expect 11.1 per cent of them to develop breast cancer – that’s 111 women.
And using our new figure, in a thousand women who have one alcoholic drink a day, we can expect 117 to develop cancer.
That’s six additional women, out of a thousand.
This is a slightly lower figure than previous estimates, but because of the weight of data behind the new 5 per cent statistic, the researchers are confident that it is a much more accurate picture of reality, rather than a statistical fluke.
What does that mean for the woman at the bar?
A 5 per cent increase in risk isn’t huge. But it’s worth remembering that this is just for one drink a day, and it all adds up – the more people drink, the higher the risk of cancer.
Other studies have found that drinking more heavily, like four small drinks (about 6 units) a day, increases breast cancer risk by 55 per cent.
Repeating our calculation above, this translates into 172 women per thousand – an extra 61 women compared to 1000 non-drinkers.
And overall, in the whole population, all these risks can add up to a lot of people. For example, our recent study on lifestyle and cancer estimated that 3,000 cases of breast cancer a year are related to alcohol.
We often write about the overall balance of evidence. And that’s exactly what this study is about. It’s a way of putting all the evidence on the scales and seeing where the needle points.
And taken together with what research has already shown about alcohol and the risk of breast cancer, this new evidence means we can be more certain that there is a risk associated with low-level drinking, but it is a fairly small one.
Our overall message remains the same – the more you cut down on alcohol, the more you can reduce the risk of cancer. And for women making choices about their lifestyle, that’s worth knowing.
Sarah Williams is a health information officer at Cancer Research UK
*Whether this assumption is valid depends on the proportion of women who drink heavily in the general population. We actually looked a range of situations using figures in between 1 in 8 and 1 in 9 for the risk of non-drinkers – the absolute figures still came out at around 6 extra women per 1000 in all cases.
- Seitz, H., Pelucchi, C., Bagnardi, V., & Vecchia, C. (2012). Epidemiology and Pathophysiology of Alcohol and Breast Cancer: Update 2012 Alcohol and Alcoholism DOI: 10.1093/alcalc/ags011
RS April 3, 2012
I totally agree with YogaSeeker’s comments!
Such a complex study would be extremely interesting.
YogaSeeker April 1, 2012
I think the study needs to go one step further in defining – what type of alcohol specifically & whether other socio-economical factors which has a direct correlation to diets etc could also have an impact on the outcome of the study..