Math clogs people’s brains

One of the most common drugs people have is Atorvastatin known more popularly as Lipitor:

Hmm…misleading?

I can’t speak for everyone else, but when I see a number like that it makes me think that you can prevent a heart attack in more than 1 out of 3 people. But four years ago, Businessweek pointed out the details of the fine print below the glaring asterisk of the 36% number.

“That means in a large clinical study, 3% of patients taking a sugar pill or placebo had a heart attack compared to 2% of patients taking Lipitor”

What that boils down to is maybe one fewer heart attack per 100 people. What wasn’t covered in the article was how such a huge number came to be. That 36% is a real number and this is how its done.

In statistics it’s known as relative risk and compares the ratio of heart attacks between both groups. I’ll just use the 2 and 3 percent numbers they provided instead of looking up the actual study. That breaks down to:

Lipitor: 2 heart attacks per 98
Sugar Pill: 3 heart attacks per 97

So the odds of heart attack for the Lipitor group is 2:98 which is 0.02041 and the odds of heart attack on sugar pill is 3:97 which is 0.03093. Next, you take these two numbers and you do (odds with Lipitor)/(odds with sugar pill) which is (0.02041)/(0.03093). This comes out to 0.66 which means the odds of a heart attack on Lipitor is 66% to the odds of having one with a sugar pill. You then subtract 1.00 from the percent: 1.00-0.66 and you get 34% lower. The reason it is not 36% exactly is because the 2 and 3 percent numbers used were probably rounded.

The confusion comes because we don’t often think in terms of relative risks but absolute risk instead. Was that Ad portraying the 36% as a relative risk instead of an absolute risk? Hmm, that depends on the reader but I definitely thought it was absolute risk when I first saw it.

BTW if sugar really is the culprit behind heart disease and not fat….maybe they shouldn’t be giving sugar pills to control groups in these sort of studies!

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s