Lots of people like to buy lottery tickets. The chance of winning is exceedingly small, but the risk is also small, is known, and is experienced up front: a little bit of money. People who buy lottery tickets make the decision that the entertainment value of thinking about winning is worth the small cost. It is a pleasant experience for those who do it.

Having a health screening test is a little like playing the lottery, but there are some important differences. Winning this kind of lottery means that you find a disease or condition that might kill you early enough to treat it successfully. If you are at average risk for the condition being screened for, your chances of winning this lottery are also very small. The risks for this kind of lottery are also small, but unlike buying a lottery ticket, the risks are uncertain and are not necessarily experienced up front. Risks include having a false positive test (the test says you have the disease but you don’t), a false negative test (the test says you don’t have the disease but you do) or complications of the test itself that may cause physical harm. You may also be over diagnosed, which means that you have the disease, but it is so small or minor that it would never harm you. Over diagnosis means that you are likely to get treatment that you don’t need, which may indeed harm you.

Take breast cancer for example. If you are a caucasian woman without any high risk factors your lifetime risk of dying from breast cancer is 2.7% (If you click this link, look at table 1.20), which means that 27 caucasian women out of 1000 will eventually die from breast cancer. African American women have a somewhat higher risk.  Mammograms every other year from age 50 changes that number to 25. That is a reduction in absolute risk of half a percent. For a single mammogram the absolute risk is reduced by only about five hundredths of a percent, which would change the 27 number to 26.95 Clearly, your chance of winning this lottery are very small.

The false positive rate for mammograms is 60 % which means that if 1000 women have mammograms, 600 of them will have abnormal mammograms that don’t turn out to be breast cancer. This is a high risk that can cause worry and psychological harm

The false negative rate for mammograms is 20%, which means that of 1000 women who have breast cancer, 200 of them will have normal mammograms. This is not a very accurate test!

The over diagnosis rate for breast cancer is 0.5%, which means that of every 1000 women who get mammograms 5 will be treated for breast cancer unnecessarily. This means that more people will be harmed by mammograms than will be helped.

I use mammograms as an example not because they have an especially high risk compared to other screening tests, but because clinicians and advocacy groups have promoted the benefits of mammograms with very little discussion of the risks. The same kinds of numbers hold true for colon cancer screening and even more for prostate cancer screening. Cervical cancer screening is a bit different. I will discuss all these screening tests in other posts.

Let’s return to the analogy of buying a lottery ticket. Suppose that in this lottery a small number of tickets contain explosives so that when you scratch one of these tickets to find your number it explodes causing serious injury. Would you participate in a lottery like that? You might, but only if your chance of winning we’re a lot higher than your chance of getting an exploding ticket.

For our health screening lottery that would be the case if you had a higher than average risk of having the disease being screened for. For example if your mother and your sister had breast cancer, then you certainly might take the risk of having regular mammograms.

For people at average risk for the disease being screened for, however, the chance of getting an exploding ticket is at least as high and for some tests higher than the chance of winning. Would you buy this lottery ticket?