The Blog

# Is This The Best Treatment For Me Doctor? Understanding Breast Cancer Statistics

You may need to make decisions about your treatment, particularly for cancer, but other illnesses too. To make rational decisions you MUST understand the numbers your doctor quotes. He won't deliberately mislead you, but statistical data can be confusing.

When we read about impressive new drugs, do we really understand what the numbers mean?

You may need to make decisions about your treatment, particularly for cancer, but other illnesses too. To make rational decisions you MUST understand the numbers your doctor quotes. He won't deliberately mislead you, but statistical data can be confusing.

So let me give you an inkling, not a maths lesson, just a few tips to help you ask the right questions.

Wow. But let's look closer: In that particular study, 95% of those who took the treatment for 15 years were cancer-free, compared to 91% who stopped at 10 years.

Hang on - 95% isn't that different from 91%. How is that cutting risk by 34%?

Well (and this is important) improvement percentages quoted in newspapers, and by doctors and scientists, are often described in relation to the original risk.

In these patients, the original risk of cancer recurring was only 9%, so any improvement would appear large relative to 9%. If the original risk had been higher, the same benefit would have appeared less.

Just to be clear, I am not suggesting that these hormone drugs are not a good thing - they have certainly helped many, many people with breast cancer. However you do need to be clear about the benefit for you, as an individual, and that means understanding the numbers quoted.

OK that's the bottom line. But for the curious, here is a worked example:

Relative Risk Reduction is a statistic often used to describe drug benefit. It is what it says - the reduction in risk (eg risk of death, or disease recurrence) relative to the original risk, ie the actual risk improvement divided by the original risk.

The Table shows an example. Patient A has a low risk of cancer returning (9%); Patient B's cancer is more likely to recur (50%).

You can see from the Table that Patient A's actual risk will only decrease by 4% with Treatment X, whereas Patient B's actual risk will decrease by 15%.

Knowing this, Patient B should be more inclined to take treatment X than Patient A.

However, if Patient A's doctor describes the benefit as Relative Risk Reduction (RRR) (see Table), then Patient A's risk appears to decrease by a massive 44%. Consequently Patient A may have unrealistic expectations for the treatment. The doctor isn't tricking him, RRR is scientifically valid, but you need to know what it means.

This example highlights another point. Sometimes we only know that a treatment works, on average, in most people. However sometimes there is more information about how much it works in different patients eg Patient B would respond to Treatment X more than Patient A.

If available, you need specific information on the benefit for you. This could influence your decision, particularly for a treatment which has significant side-effects.

So, the final bottom-line: