The latest Fat Head Report video is about yet another “meat causes colorectal cancer!” study a reporter for the Telegraph described as “frankly terrifying.” It’s such weak observational nonsense, Walter Willett would no doubt approve.
Here’s why you don’t need to be terrified. Transcript below the video.
Hello, I’m Tom Naughton and this is the Fat Head Report.
Well, I guess I need to create an ongoing segment called Meat Will Kill You.
Meat won’t kill you of course, but there are people who desperately want you to believe it will. So they’re constantly producing studies designed to scare you.
And when I say scare you, I’m not kidding. A writer for the UK telegraph described the most recent meat will kill you study as frankly terrifying.
So let’s take a look at the reason this writer for the telegraph was apparently hiding in her closet, in case her home was invaded by several ounces of murderous meat.
Oxford University research on half a million people found that eating red meat just once a day increased the risk of bowel cancer by a fifth.
Wow, that IS terrifying. Unless you understand how these studies are done.
A study like this is what’s called an observational study. And to explain what that means, here’s a clip from my recent film, Fat Head Kids.
If you conducted the study Dr. Fishbones did, you wouldn’t have to guess who does or does not have a tattoo. So at least you’d be starting with accurate data.
For dietary studies, that’s not the case. To determine what people eat, researchers have them fill out a food survey that looks something like this. Do you see the problem here? These surveys are wildly inaccurate. Some people report eating so little, they couldn’t possibly develop cancer, because they’d starve to death first.
Then near the end of the study, researchers have people fill out the same survey again. And from this, they decide what people have been eating for the previous five or 10 or 20 years.
But let’s suppose people can actually remember how much of everything they been eating and report it accurately. You still have that problem with the confounding variables Mr. Spot mentioned.
And guess what? In this study the people who developed colorectal cancer didn’t just eat more meat. They were also more likely to smoke. They also drank more.
Ah, but wait, the researchers tell us, we adjust our data for lifestyle factors like smoking and drinking.
Well, actually, they adjust the data based on how much people tell them they’ve been smoking and drinking.
But heavy drinkers routinely underestimate how much they drink. And smokers routinely underestimate or just plain lie about how much they smoke – or even if they smoke.
But okay, let’s forget all that. Let’s pretend that by some miracle, the researchers had accurate data all the way through, top to bottom.
In that case, would the higher cancer rate among the heavy meat eaters be frankly terrifying? After all, one-fifth more colorectal cancer sounds like a lot, doesn’t it?
Well, let’s look at the data. This was a study of nearly half a million people.
Among the people classified as labeled as lower meat eaters, 5.49 out of every thousand developed colorectal cancer.
And among the people classified as heavy meat-eaters, it was 6.58 out of every thousand.
So for the people classified as heavy meat eaters, the actual increase in the odds of developing colorectal cancer was one in a thousand.
And here’s the final reason not to be terrified. Good scientists don’t accept a hypothesis unless the evidence supporting it is consistent.
So if we say this food or this behavior causes cancer we need to see that result over and over and over.
Which is the case, for example, with smoking and lung cancer. You’re never going to see a study where smokers and non-smokers have identical rates of lung cancer. And you’ll certainly never see a study where the smokers have lower rates of lung cancer.
But in this observational study, people who ate red meat did not have a higher rate of colorectal cancer.
And in THIS observational study people who ate more meat did not develop more colorectal cancer.
And in this observational study it was the vegetarians who had a higher rate of colorectal cancer — 39 percent higher.
If eating meat causes colorectal cancer, how can that possibly be?
Good scientists don’t just pick and choose the results they like. And when it comes to meat and colon cancer, the results are all over the place. Which means it’s highly unlikely that meat causes colon cancer.
So enjoy your bacon.
And you can stop screaming now.