Think maybe you should be on statins because of your high cholesterol level?

They're wonder drugs, aren't they, lessening by, say, 31% the likelihood you'll keel over and the next thing you know somebody will be brandishing two shiny disks over your chest and yelling "Clear!"

Well, no. As this Slate piece, written by a real heart doctor, notes, you have to know the difference between a relative reduction, an absolute risk, and a far more useful measure of drug effectiveness called NNT, or number needed to treat. That 31% figure translates into only a drop of only 2.2 percentage points, from 7.5% to 5.3%, in terms of absolute heart attack risk.

Out of 100 people taking a statin drug, perhaps two would have avoided having a heart attack because of it. So the NNT is 50. Aspirin, that wonder drug, has an NNT of 208. Higher is less useful.

The NNT is intuitive: To a savvy, healthy person with high cholesterol that didn't decrease with diet and exercise, a doctor could say, "A statin might help you, or it might not. Out of every 50 people who take them, one avoids getting a heart attack. On the other hand, that means 49 out of 50 people don't get much benefit."

This is not to say that smoking drinkers with family histories of heart attacks and blood as thick as glue should throw away the pills; they should not. But it is to observe that drug companies would rather quote the relative risks than the absolute risks or the NNT. It's a bigger, scarier number.

And to note that people who write about fantastic new drugs for this and that do readers a disservice by first not addressing their "innumeracy and scientific illiteracy", as the writer of a famous book on the subject terms it.