Deciphering studies: Absolute vs Relative Risks

I thought this example Lisa Martinez made in the comments at the Well blog was one of the clearer examples I’ve seen for absolute vs relative risks.

100 women took ABC pill and 100 women took a placebo, which is not an actual medication. Of the 100 women who took ABC pill, 2 developed cancer and of the 100 women in the placebo group, 4 got cancer. It is reported that this clinical trial showed a 50% reduction in cancer and another report states that the same clinical trial showed a 2% decrease in cancer. Both percentages are accurate and that is because the data are being reported in two different ways.

The 50% reduction was reported using relative risk reduction. The 2% decrease was reported using absolute risk reduction. It is extremely important that you understand the difference between the two when making decisions about your care. So when presented with options for chemotherapy or any treatment, make sure you ask your healthcare provider to give you the relative and the absolute risk reduction percentages.

Many publicists and news articles on the risks of weight use relative risks not absolute. As you can see, this can be a BIG difference!

7 thoughts on “Deciphering studies: Absolute vs Relative Risks

  1. Great example. Here’s how I understand it, and please correct me if I’m off base here:
    Relative risk is determined when the experiment/study involves those that are being treated for the same disease/disorder/etc. As in, the participants could be viewed as relative to each other because they had something in common?
    Absolute risk is determined by factoring the data from the study participants against the rest of the population? Study participants vs. everybody?
    If this is true, then I finally understand how the data can seem so vastly different! But what does it all mean? Logically, when factoring your situation (treatment options) against the entire population (however it is broken down–gender/age/body type), I would expect that the percentage would be lower because, well, that’s a lot of people. If I were wanting to decide whether or not to choose a specific treatment, would it then be wiser to ask about the relative risk and make the decision based on that data? It will be interesting to know which data most doctors inform patients with.

  2. Actually Regina, maybe I’m reading your comment wrong, but it is a little different than that. Lets say the CDC says that 10% of people in the US have the zombie virus, and 20% of fat people in the US have the zombie virus. A reporter could say that 10% more fat people get the zombie virusthan thin people, OR they could say fat people are twice as likely than thin people to have the zombie virus.

    It is just about how they compare the percentages. It’s either X-Y= Absolute Risk or Y/X= Relative risk, and the relative risk often sounds much scarier. (this may actually be what you said….but makeup of the populations doesn’t really matter as much.)

  3. To Regina – You don’t quite have it – note all numbers following are made up, just examples. In any group of 100 women, 4 will get breast cancer, this is a 4% absolute risk. Now lets say you do something to a group of 100 women and only 2 of them get breast cancer – this group has a 2% absolute risk of getting cancer. Now if you compare the two groups, that’s where relative risk comes into the picture – since if you do nothing 4% will get cancer, and if you do something 2% will get cancer – that means that 50% of those expected to get cancer, will not. So relative risk is comparing to what’s expected.

    Now, where this is very important is lets say the ABC drug, while reducing your risk of breast cancer, increases your risk of a fatal heart attack. So the drug makers say reduces cancer by 50% and increases heart attack by 3% – sounds good, until you ask questions and find out that one figure is relative and the other is absolute. So back to our 100 women. They take the drug, 2 of them don’t get cancer – that’s great. Unfortunately, 3 of them get fatal heart attacks. So, with ABC drug, 3 people die of heart attack and 2 people get breast cancer. Without ABC, 4 people get breast cancer. These numbers certainly change how you feel about ABC drug, do they not?

  4. Pingback: August at Living 400lbs… | Living ~400lbs

  5. I am so glad to see this discussion of relative risk vs absolute. Drug companies use (and abuse!) the data from their studies that show their products in the best light.
    Dr. Nortin Hadler makes this exact point in his great books, Worried Sick and The Last Well Person. He also wrote this: If There is No Benefit, Why Tolerate Any Risk?
    Naturally, he got several snarky comments after his post, but nobody refuted his facts.

  6. Pingback: Cool Shit 3 — « Fierce, Freethinking Fatties

  7. Pingback: But science can be used to prove anything! « Zaftig Zeitgeist

Leave a Reply

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

You are commenting using your 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