Putting risk in 'real' terms
To many people, saying risk factor 'x' increases your risk of 'y' cancer by 'z' per cent is quite meaningless. So it is usually better to put risk in terms of numbers of people affected. This can be best demonstrated with an example.
In 2003, the Million Women Study, funded by Cancer Research UK, found that taking certain forms of hormone replacement therapy (HRT) increased breast cancer risk in post-menopausal women 'by two per cent'. This is a measure of absolute risk. But a much better and clearer way of communicating this research is to say:
"If we follow the health of 100 women, aged between 50 and 60, who do not take HRT, we would see that two women, on average, get breast cancer over five years.
When we look at 100 women of the same age who DO take combined HRT, we find that four women, on average, get breast cancer over the same period."
These are the same data - an absolute increase in breast cancer risk of two per cent - but presented in a much more understandable way.
Many newspapers and media sources reported this increase as a 'doubling' of risk. This is absolutely true - an increase from two women to four per cent is a doubling of absolute risk. But to say that a person has 'double' the risk of cancer has much more impact than saying a 'two per cent increase'. This reflects different agendas. News media outlets have a tendency to emphasise the dramatic; health agencies will have a more cautious approach.
Scientists also communicate risks by comparing them to other, well-understood risks.
Let's look, for example, at the increase in breast cancer risk due to alcohol on the previous page.
The absolute increase in risk here is about 1.4 per cent. This means that, for a hundred women who drink two or three units of alcohol per day, we would expect about 1.4 extra woman per hundred to get breast cancer, compared with a hundred women who were tee-total. We can compare this to the risk due to combined HRT use (above), which causes an extra two cases of breast cancer per hundred women.
So HRT is a more important risk factor for breast cancer than moderate alcohol consumption.
