It is a good morning exercise for a research scientist to discard a pet hypothesis every day before breakfast. It keeps him young.
It’s science today, folks. If you want to run screaming for the door, now’s the time.
I noticed two headlines in the news today. One illustrates total bias in the way something is reported; the other shows proper scientific reporting of a finding that may generate a hypothesis worth looking at. The first is designed to frighten you, the second is designed to inform you.
One of the things that led me to the healthy path I’m now on was someone–in my case Gary Taubes–opening my eyes to the manner in which health “news” is reported. Reminding me of the difference between a hypothesis and a theory and the incredible scientific importance of not treating a hypothesis as anything more than it is until such time as it has been rigorously tested. Illustrating how the old scientific method of formulating a hypothesis and then setting out to prove it wrong by every method available in the interests of truth has been largely abandoned in the field of nutrition, replaced with assumptions that hypotheses are correct and vicious resistance to any true test of their merit.
So let’s look at these two headlines. The first is:
The second is:
This is probably going to take two posts, so…sorry about that.
Let’s start with the first article, shall we?
It claims that eating red meat was associated with a “48 percent increased risk of developing” diabetes. We are told the “association was greater” with processed than unprocessed meats, but we aren’t told the numbers for unprocessed meat. We’re just left with the one big scary number.
Before anything else, let me just tell you this: this study is worthless. It relied on self-reporting, which is nose-holdingly notorious in its unreliability, and the percentages they found were so small as to be within the range of statistical noise.
But even if it’s not statistical noise, and even if everything in this study was absolutely true…so what? Don’t let that big, bad 48% number fool you. They are, deliberately, not telling you the difference between relative and absolute risk. 48% sounds really horrific, and they know you’re assuming absolute risk when you read that. But the absolute base risk for the study’s participants was only around 8%. The processed meat eaters experienced a 48% relative increase, meaning their actual absolute risk was still less than 12%. That’s 12% for the smoking, drinking, carb-loading, sugar-guzzling, transfat-snorking meat eaters, and 8% for everyone else: including the vegans.
They conveniently forgot to mention the risk in relation to fresh meat eaters, so I’ll do it for them. The last time this analysis was done, in 2011, they came up with a 51% relative risk increase for the processed meat eaters, and only a 19% relative risk increase for the fresh meat eaters. So that would make the absolute risk 8% for the average person and 9.52% for people who ate a whole lot of fresh meat.
If the difference between me and my sirloin and Acacia and her lifetime of soy burgers is a measly 1.52%, then pass the A1 sauce, please. (More on relative risk here.)
But if you’re still worried about this study, let’s consider some of the details.
In the first place, this is a study in the same sense that the Democratic People’s Republic of Korea is a republic. You can call it a republic all you want, but until you get everyone out of the prison camps no one will take that too seriously.
This “study” is actually just an analysis of the data from three studies done some time ago. And it’s nothing new. Almost the exact same analysis was done back in 2011.
So back in 2011 you see an association, which by itself is utterly worthless (especially when based on unreliable food surveys) except to point you in a possible direction for a hypothesis. So would your next logical step be to tell everyone to stop eating red meat?
No, your next step should have been to postulate a hypothesis based on the association you saw: something in red meat causes diabetes. Then you should have set about trying to disprove it with controlled trials. You could also set about trying to explain it by investigating the physiological mechanism by which the meat could possibly be causing the diabetes.
Then, when you kept getting consistent results and couldn’t disprove it, you could start calling it a theory and warning people about eating red meat.
But that’s not what these people did. No controlled trials or investigation of any of their wild speculations about how on earth meat–which doesn’t affect blood sugar–could possibly be causing diabetes. Nobody did that. Instead, they simply waited a few years, repeated the same analysis of the same material, and then trumpeted it as if it was all brand new, earth-shattering information that absolutely proves the only safe thing for you to do, Tubby, is put down that ribeye and get yourself some Tofurkey.
And what about that 2011 data analysis? There’s some serious problems in there.
For one thing, the people in the highest quintiles for meat consumption were also in the highest quintiles of some other things: smoking, drinking of both alcohol and soda, total calories consumed, weight, trans-fat intake, potato intake, coffee intake, and refined flour intake. They were in the lowest quintiles for exercise and intake of fruit and vegetables. These were people who didn’t care about their health.
Now the brilliant data analysts will tell you that they “controlled” for all that stuff. They were somehow able to wave a magic wand and make themselves omniscient, thereby completely and thoroughly understanding each and every effect on every individual of 1. too much alcohol, 2. too much soda, 3. too much tobacco, 4. eating more calories total, 5. being fat, 6. eating lots of trans fat, eating tons of starch in the form of 7. potatoes and 8. refined white flour, 9. taking in more caffeine, 10. never exercising and not eating 11. fruits and 12. vegetables. They also would have to know all the possible effects of every combination of those 12 that might effect any given individual. With their newfound omniscience they decided to isolate meat as the factor that was causing this statistically insignificant increase in diabetes risk.
Oddly they seem not to have chosen to use their new powers to end world hunger or find Amelia Earhart.
Another problem is that they didn’t bother to account for some variables at all. Added sugar, for one. They counted the sugar in soda and Little Debbie snack cakes, sure. But they didn’t account for the sugar Fat Jack was dumping in his coffee or sprinkling on his raisin bran every morning. I know several people that put upwards of 5 tablespoons of sugar in just one cup of coffee or tea in the morning. Not counting that is downright dishonest.
Then they didn’t bother to consider any difference between a steak and cheese footlong at Subway, and an actual steak. Since most people, other than me and hopefully you, eat their meat–especially their processed meats–with plenty of carbs, this means it’s pretty much impossible to isolate the different impacts even the processed meat could have from the impact of the rolls, french fries, mashed potatoes, buns, and bread.
Finally, when looking at ratios of different fats, they only counted the fat in actual whole foods, like eggs and cheese. They didn’t bother to count the oil in which anything might be cooked. You know, like the oil that french fries are fried in.
So all we have is some data cobbled together from unreliable self-reporting food surveys, with confounding factors never accounted for, comparing people with vastly different lifestyles in every respect. From this, apparently, we are supposed to be frightened into cutting down on the meat by a statistically insignificant association in risk for getting diabetes.
This analysis that came out today is the same data with all the same problems. No one even bothered to investigate the laughable hypotheses put forward to explain this association. In fact, no one seems to have found it at all important to get a “why” at all.
Let’s deal with those hypotheses, shall we?
I shall randomly pick the first one to be first. It’s also my favorite:
Several mechanisms may explain the adverse effect of red meat intake on T2D. First, the association may be mediated through the effects of heme-iron derived from red meats (33, 34). Iron is a strong prooxidant that catalyzes several cellular reactions in the production of reactive oxygen species, and increases the level of oxidative stress (35). This can cause damage to tissues, in particular the pancreatic beta cells….
Wow. Sounds scary, eh?
Beef is the meat with the most iron by far, and a 12oz ribeye has about 35% of your recommended daily allowance. You’d have to eat three a day just to get yourself to a little bit more than what you need, and very few people are eating that much meat. I’m a heavy meat eater by anyone’s standards, and I’d be hard pressed to eat a whole 12 ounces of beef in a whole day. Unless you’re chowing down on enormous portions of liver all day, it is highly unlikely you are overeating iron.
This is how you get too much iron:
Especially if you’re combining it with this:
And let’s talk for a second about oxidative stress, shall we?
What they’re saying is–very, very basically–that iron and oxygen can “rust” you. This is very true.
Reactive oxygen species (ROS) and advanced glycation endproducts (AGE) are indeed toxic. They are the essential elements in this human process of “rusting.” ROSs form when an electron attaches to an oxygen atom in such a way that it transforms the atom from something benign into something that can react chemically with other molecules. We call them “oxidants” because they oxidize other molecules. Nutritionists want you to eat lots and lots of fruit-antioxidants!-to counter these.
Only you wanna know how those are actually formed? ROSs form through the process of burning glucose for fuel. Yes, that’s right, like the glucose you get from your whole wheat toast, whole grain breakfast cereal, vegetable juice, and…wait for it…fruit. (Or your doughnut, soda, milkshake, and Cocoa Puffs, if you’re that kind of eater.)
Now ROSs are perfectly natural byproducts, and they even have some important uses in the body. The problem only arises when you are burning nothing but glucose for fuel all the time and consequently have an overabundance of them damaging your DNA. (If you like, read more about ROSs here, here and here. And here, too. And here!!)
How about AGEs? These begin when your blood sugar is high. In the wild time of that sugar high, a sugar molecule makes an ill-advised marriage with a protein; but the proper enzyme to control that marriage and make it happy is absent. There are simply not enough enzymes around to chaperone all that sugar. Without the enzyme, the marriage goes sour and there’s just screaming and plate throwing, and eventually the protein goes looking for Someone Else and starts getting stuck to other things that it was never meant to stick to. Then all the parties in this love triangle start reacting and rearranging until you get AGEs, which then rush down to the Dysfunctional Marriages Anonymous meetings and meet other AGEs and some random proteins that are cruising around looking for emotionally vulnerable partners, and the cross-linkage causes serious cellular damage.
The good news with AGEs is two-fold. First, if you put down that doughnut (or, let’s face it, that “healthy” fruit smoothie) and stop filling your body with excess sugar all the time, the AGEs formation process has a hard time even beginning. Second, even if they do begin–after the Thanksgiving buffet, perhaps–lowering your blood sugar again causes the sugar and protein to sober up and get an annulment before anything too serious happens.
The bad news is that if you keep your blood sugar chronically elevated, these bad marriages just keep forming, and you’re in for some serious misery like diabetes and cancer. (If you like, you can read more about AGEs here, here or here.)
Too much iron, my foot.
Their second idea is this:
Nitrites and nitrates are used frequently in the preservation of processed meat, and they can be converted into nitrosamines through interaction with amino compounds either in the stomach or within the food product. Nitrosamines have been shown to be toxic to pancreatic beta cells and to increase the risk of diabetes in animal studies.
This would be funny if it weren’t so sad. The conclusion of their study is that you should cut down on all red meats. But now they’re trying to explain it based on something that you only find in processed meats. If they really thought nitrates were the problem, they wouldn’t need to tell you to avoid fresh steaks.
Not only that, but the primary dietary source of nitrates is green vegetables. There’s more in celery than in bacon, and yet you don’t see anyone warning vegetarians that they should restrict their green vegetable intake. Arugula contains 4,677ppm of nitrate. A hot dog has 10ppm. Celery has 1,103ppm. And your own body produces so many nitrites that the amount you could possibly consume from food is insignificant. (More here and here.)
That’s what I mean when I say their bias is showing. They don’t want you eating red meat. They already decided that before they did the “research,” and the reporter had already decided it before he wrote about it. The research itself, and the report about it, were carefully crafted to fit the conclusions they wanted to find.
I’m going to stop here and in the next post examine the second article I mentioned. That article is an example of true scientific reporting and worth comparing with the nonsense we’ve talked about here. But if you’re going to take anything away from this post, let it be this: excessive sugar, not fat, is your enemy; even when it sneaks across the border dressed in healthy whole grain camouflage.
Plato says he’s hungry
Help us keep paying for this site and feeding the dogs.