Risks and Benefits: Understanding the Statistics that Affect You

Health Guide
•

One of the first things I did once the shock of my breast cancer diagnosis wore off was start to worry about what my diagnosis meant for my two sisters and my daughter.   My doctor told me that their risk had increased by 50%.

How could I tell them that they had a 50% chance of getting cancer?  With three first degree relatives to worry about and a fifty-fifty chance for breast cancer, it seemed inevitable that we would soon be a family with two or more breast cancer patients.

Fortunately, before I called them, I did some more reading and learned that I had made a common statistical mistake.  I had confused absolute risk with relative risk.  To know my relatives' real risk of breast cancer, I needed to understand their initial risk and add 50% to that.  (The 50% figure my doctors gave me was based on my type of cancer and my medical history.  Only your own doctor can tell you how your diagnosis affects your family members' risk.)

In the United States about one in eight women will have breast cancer sometime during her lifetime, so a better way to think about my daughter's and sisters' risk would be to take their lifetime risk of about 12% and add 50% of 12% to get a risk factor of about 18%.  That's a huge difference between thinking that the odds were even for getting breast cancer or staying well, and realizing that there was a less than 20% chance they would get it.  I could imagine a room with five women with the same risk as my relatives and see that four of them would be just fine.

Dr. Susan Love in Dr. Susan Love's Breast Book points out that I made another statistical error in my estimate.  The one in eight number is the life-time risk for all women, and it includes women who have many risk factors.  Dr. Love says that what I should have been starting with is the 3.3% risk for women with no risk factors and adding 50% of that.  Since my family members do have some other risk factors, I'm comfortable with my down and dirty stats.  The important thing to me was that I went from assuming that it was almost inevitable that at least one of them would get breast cancer to realizing they had a better than 80% chance of staying well.

I had to learn to do the same kind of calculations when making treatment decisions.  If the doctor tells you that a certain treatment reduces your risk of recurrence by 50%, that number is meaningless unless you know what your original risk of recurrence was.  If your original recurrence risk was 80%, a 50% reduction means that your risk of recurrence is now 40%.  That is a huge difference, and it would be worth taking a difficult course of treatment for those results.

However, suppose your risk of recurrence without that treatment was 10%.  A 50% reduction would be 5%.  Your risk of recurrence would go from 10% to 5%.  That's significant too, but if the course of treatment is difficult and has potentially dangerous side effects, you might decide you'll stick with your 10% risk and deal with recurrence if it happens.

•