Over the last week, I’ve been reading a terrific (yet terrifying) book.
It’s called Risk Intelligence, and it talks all about our ability to estimate the probabilities of things.
Sounds nerdy, I know.
So let me give you a specific example of something they talk about:
Which group do you think is better at making better predictions?
B. Weather Forecasters
Think about it for a minute…
OK–what’s your answer?
If you said A, doctors, you’re *wrong*.
From the book:
“When the weather forecasters in the first study said there was a 90 percent chance of rain the following day, it rained almost 90 percent of the time. But when the doctors in the second study estimated that there was a 90 percent chance that their patients had pneumonia, only about 15 percent of the patients turned out to have the disease. In other words, the doctors had much more faith in the accuracy of their diagnoses than was justified by the evidence. That meant they were likely to recommend more tests than were strictly necessary, prescribe more treatments than warranted, and cause their patients needless worry.”
There are a couple of reasons why this is the case.
Number one: Doctors don’t get good feedback.
After they give a diagnosis or recommend a treatment, they rarely get good, clean information about how the treatment went. This is particularly true for those who work in hospital settings–since there are no long-term relationships with patients.
Number two: Most doctors are forced to make decisions about unreasonably wide a range of problems.
Becoming well-calibrated in a particular area requires *a lot* of feedback, a lot of trial and error.
However, most doctors have to make diagnoses across a range of bodily systems (lungs, arms, hormones, etc.). They’re a mile wide and an inch deep.
Thus, they never really gain the expertise necessary to make accurate predictions in a specific area.
I’m not saying that you shouldn’t trust your doctor… but you should stay vigilant. They’re human, after all… and they’re *way* less accurate than weather forecasters.