Imagine a highly reliable cancer test. It detects 95% of a certain type of cancer, and has a "false positive" rate of only 1%. This test is used on a population in which this type of cancer occurs in 0.5%. One day your doctor tells you that you have tested positive. What is the chance that you are actually sick? Surprisingly, it is only about 32 percent! The mathematics of health screening (and of epidemics as well) are not necessarily intuitive. Opposition to the recent evidence-based report recommending less frequent mammograms needs to be supported by facts. We need a health care system based on science rather than emotion.
John Allen Paulos