Do You Really Know Your Health Risk?
Odds Are You Don't -- And Neither May Your Doctor
Sept. 25, 2003 -- Your life depends on knowing your health risks. Yet a new study shows few people -- and surprisingly few doctors -- really understand risk.
It's not that the math is too hard. The problem is that the way doctors talk about risk clouds people's minds, says Gerd Gigerenzer, PhD, director of the Center for Adaptive Behavior and Cognition at the Max Planck Institute for Human Development, Berlin.
"Risk is often expressed in ways that cloud patients' minds," Gigerenzer tells WebMD. "It clouds the mind so that patients don't even notice they don't understand their true risks."
Gigerenzer is one of several scientists whose reports on health-risk communication appear in the Sept. 27 issue of TheBritish Medical Journal. Their consensus: Doctors aren't doing a very good job of explaining health risks to patients.
True, lots of people are mathematically illiterate -- innumerate, as Gigerenzer puts it. But that's no excuse, he says. Just being bad at math doesn't mean you can't understand risk.
"It is not a problem with people's minds," Gigerenzer says. "The key problem is with how information is presented. The problem is in the form, not the content."
What's My Health Risk?
Let's say taking a certain pill will cut your risk of X disease by 50%. Half the risk! That sounds great. Give me that pill!
But what if only two out of every million people get disease X? Halving you risk to one in a million doesn't sound nearly as good. And if the pill has side effects, it might do more harm than good. So should you take the pill? You might, depending on how scared you are of disease X -- and how you feel about risking side effects. Your choice -- even having a choice -- depends on all this information, not just on the "cuts risk by 50%" figure.
Most doctors wouldn't be fooled by this example. But here's one that few doctors get right. The scenario looks like this:
- Suppose the risk of colon cancer is 0.3%
- A test detects colon cancer in 50% of people who have the disease
- 3% of the time, the test says a person has colon cancer when he doesn't
What is the chance that someone who tests positive actually has colon cancer?
Doctors are supposed to know this stuff. But more than half of them were way off, guessing 50%. The correct answer is about 5%. Answers ranged from 1% to 90%.
"Doctors don't get the right training for communicating health risk," Gigerenzer says. "Many young people interested in science study medicine because they want to avoid statistics and psychology. And even if they get statistical training, they get it in a form they do not truly understand or easily forget."