Bad Science Reporting

PBS reports “Bacon, hot dogs, and processed meats cause cancer, as dangerous as smoking”. The article itself isn’t entirely terrible, but the headline is really poor.

To get some things out of the way:

  • This isn’t really news, the WHO has rated processed meat as problematic for a few years.
  • You probably should eat less red and processed meats anyway, if you are an American.
  • Yes, it’s pretty likely that red meat contributes to cancer rates.
  • Yes, it’s pretty likely that processed meat contributes to cancer rates.
  • No, this is probably not a case of “science is telling me something else is bad and they’ll change their minds in two years”

First note: the WHO studies international health data.  The overall risk of any potentially hazardous human behavior that the WHO studies needs to take that into account.  Health standards vary by country, and what might be very hazardous in one country… say, drinking raw milk in sub-Saharan Africa… might be fairly safe in another, like the U.S.  So there’s that. In the case of “processed meats”, you have a particularly wide variation of definitions.

But we’ll let that pass for the moment…

Let’s take a quick gander at some numbers that are relevant to the United States. Lung cancer has the third highest mortality rate of major cancer types. You have an 83.4% chance of shuffling off your mortal coil with lung cancer within five years of your diagnosis. Colorectal cancer, on the other hand, has a five year mortality rate of 33.5%. This alone is a pretty significant difference.

About 8% of lung cancer is caused by inherited factors, compared to 20% of colorectal cancers.  This means that direct comparisons of these two cancer rates in randomized populations, while great for medical research, will give you a limited amount of perspective, as an individual who knows your own medical history, because genetics as a confounding factor is very different for the two cancer types.  If you know that you don’t have a genetic predisposition towards colorectal cancer, that makes you very different from a random person on the street lacking that data bit.

The base rate is significantly different for the two types of cancer. You have about a 40% chance of getting some form of cancer at some point during your lifetime. You have slightly less than a 5% chance of developing colorectal cancer at some point during your lifetime, and a 7.5% chance of getting lung cancer (more on that in a minute).

So lung cancer has both a higher base rate, and a much higher mortality rate than colorectal cancer.

According to the report, processed meats contribute about a ~18% increase in your colorectal cancer risk if you eat 50 grams a day. Since your base rate of colorectal cancer is 4.34%, your adjusted risk would be 4.34% * 1.18, or… about 5%.  That doesn’t sound too bad, does it?  That’s assuming you eat a couple of slices of bacon every day.

On the other hand, if you smoke, you’re 15 to 30 *times* more likely to get lung cancer alone than a non-smoker. In the U.S., that factoid is estimated at an increased risk of 25 times. Studies in other countries break down lung cancer risk by smoking prevalence.

In one European study, for example, researchers found that your actual cancer rate for non smokers was under 1%, whereas the cancer rate for smokers was up to 16%. Meaning… really, you have very little chance to get lung cancer at all if you don’t smoke, and a fairly significant chance of getting lung cancer if you do smoke.  Smoking also contributes to esophageal cancer, lip and mouth cancers, and other types of cancer.

Not to mention the fact that smoking is also a major contributor to heart disease, lower respiratory diseases, and stroke, as well as cancer, and that these four factors are four of the top five killers when it comes to human beings, responsible for a whopping 1,474,169 deaths in year in the United States. Smoking is terrible for you in lots of ways.

That doesn’t make bacon anywhere near “as dangerous as smoking”.  Even if they are rated in the same class of carcinogens.

More here and here and here.

(Image credit: The Culinary Geek on Flickr, Creative Commons)

Updated (13:29 26-Oct-2015) to add:

On the upside, we are starting to see an increasing amount of explanatory reporting when bad science reporting hits the Internet!  There is hope for the future.

Staff Writer
Home Page 

Patrick is a mid-40 year old geek with an undergraduate degree in mathematics and a master's degree in Information Systems. Nothing he says here has anything to do with the official position of his employer or any other institution. ...more →

Please do be so kind as to share this post.
TwitterFacebookRedditEmailPrintFriendlyMore options

42 thoughts on “Bad Science Reporting

  1. Duuuuude,

    We should be ENCOURAGING people to eat this stuff. They die sooner. That combined with limited medical treatments to folks who have these cancers, ’cause it’s their own fault for not eating healthy, we’ll save MASSIVE amounts of cash which we can then use to cover more people so they can get dental and eye care from the ACA. See how it all works?!


  2. Nuance doesn’t get enough eyeballs. It always has to be “FISH YOUR PLEASURE!! EAT NOTING BUT SPINACH OR YOU WILL DIE MISERABLY!!”

    Eh, I don’t deny the unhealth of the average diet, or of mine. Though when I was doing as close to the right things as I’ve ever done is when I ended up having to fight off the C. I’ve since decided that if DNA is that unpredictable, I’m just going to enjoy what I have left and not care. Death is going to suck anyway, why should life suck too?


  3. I take as a baseline assumption that any journalism that involves percentages, and especially comparing percentages, will be a complete botch. Because math is hard, don’t you know? We can’t be expected to keep straight what we are comparing.


Comments are closed.