Understanding Medical Statistics – A Layman’s Guide

Medicine is a science of statistics and nowhere is this more apparent than in cancer treatment.

Here is a guide to understanding some of the statistical terms you will run into when discussing cancer treatment.

Firstly, you need to know that no one can predict with total certainty any patient’s exact outcome. Instead, cancer mortality rates are based on the outcomes of large numbers of people who have had the same disease. In other words, the statistics that will be quoted to you are only probabilities.

A second thing to know is that not all cancers have the same outcomes. In other words — some cancers are deadlier than others.

And the third thing to realize is that the probability of being cured depends very much on the stage of your cancer.

For example, in all cancers, patients with stage 1 disease have better survival rates than patients with stage 2 disease. And patients with stage 2 disease have better survival rates than patients with stage 3 disease. And stage 3 patients have better survival rates than stage 4 patients.

That’s why it’s so critical for you to know what stage of the disease you have.

When doctors quote the probability of survival, they don’t tell you that you’re supposed to live 3 years and 3 months. Instead, they quote you a figure that’s called the 5 year survival. This number represents the probability that someone with your disease will be alive at the end of 5 years.

For example, if the 5 year survival is 25%, this means that the odds of being alive after 5 years are 25%. It’s important to look at the particular group of patient for whom the 5 year survival is being quoted. For example, it could include all patients with a certain type of cancer or it could only include patients with a certain stage of that cancer.

For example, 49 percent, or about half, of people diagnosed with early-stage lung cancer live for at least five years after diagnosis. So, the five year survival for early-stage lung cancer is 49%. Contrast this with the five year survival rate for people diagnosed with lung cancer that has spread (metastasized) to other areas of the body which is only 3 percent.

So, as you can see, you want to find out what the survival rate is for patients with the same stage of disease.

Remission refers to a shrinking of a cancer. Remissions can either be complete or partial depending on whether or not any evidence of cancer remains. Obviously, if no cancer can be found after treatment then the remission is complete.

If the cancer is smaller but hasn’t disappeared completely then the patient has experienced a partial remission.

A complete remission may represent a cure. But there is still a chance that the cancer will return depending on the type of cancer. In people who achieve only partial remissions, the cancer nearly always regrows.

Are there any other statistical terms I should be aware of?

The 5 year survival rate tells you how many people are alive at the end of 5 years but it doesn’t tell you how many of these people are in complete remission at the end of 5 years (in other words how many people have survived for 5 years and have no evidence of cancer).

So here are two more specific terms:

This is the percentage of people who are not only alive after 5 years but are in complete remission.

This is the percentage of people who are alive after 5 years but who still have evidence of cancer, although the cancer isn’t progressing. This includes people who may have had some success with treatment but not enough to completely eradicate their cancer.

Two other statistical terms that you should be aware of are relative risk and absolute risk:

This is the absolute difference in results between alternatives. So, for example, if treatment A increased survival by 22% and treatment B increased survival by 20% then treatment A resulted in an absolute benefit of (22%-20%) = 2%.

This is the relative difference in results between alternatives. So, in the above example, the relative benefit is (22%-20%)/20% = 10%. This number is calculated by taking the difference in outcome as a percentage.

You should beware that often results are presented as relative risk reduction or benefit because the numbers sound more impressive. For example, in the above example, a relative benefit of 10% sounds more impressive than an absolute benefit of 2%.

Hopefully, now you’ll be able to be a little more discerning about cancer statistics.