Red Meat Causes Diabetes! (This, From the ADA)

So the American Diabetes Association released the news this week that a new study suggests that red meat may increase the risk of developing diabetes. Have I panicked, and purged my freezer of all red meat? Hardly.

First of all, this is an observational study. For an interesting look at the concept of observational studies, read what Dr. Mike Eades has to say on the subject. If you don't have the time, I'll summarize: Observational studies are a decent way to generate hypotheses for further study, but in and of themselves prove exactly nothing.

Secondly, the reports can't seem to work out whether they want to dog on red meat in specific, or low carbohydrate diets in general. The article I linked to starts: An Atkins-style diet that emphasizes animal protein may actually be linked to higher incidence of type 2 diabetes, according to a large observational study. But just three paragraphs later, it clarifies, The main drivers were red and processed meat and the heme iron found in red meat... This pings my Spidey-sense, and makes me suspect that the whole point of the study was to undermine confidence in low carb diets in general, not to implicate red meat in particular.

Third, notice how they've lumped processed meat in with red meat? Studies always do that, for some reason. We're supposed to believe that a steak is nutritionally equivalent to, say, bologna or Spam. The studies "demonstrating" a link between red meat consumption and colon cancer do the same thing -- but when you take out the processed meat, the link disappears. We won't even talk about whether grass-fed steak and feedlot steak are nutritionally equivalent, since so many low carbers can't yet find or afford grass-fed beef.

But those are minor quibbles compared to my two big questions:

Did the low carbers self-select their low carbohydrate diet? Because if they did, as I strongly suspect is the case, then it seems to me that the chances are excellent that the low carbohydrate group included a disproportionately high number of people who had already self-diagnosed a faulty carbohydrate metabolism. In other words, those of us who looked at our big guts, our tendency to be starving by 90 minutes after those Cheerios in the morning, and our wildly fluctuating energy levels and said "Hmmm. Maybe carbs aren't my friend," are more likely to have been prone to diabetes in the first place. Indeed, many of us may already have been undiagnosed diabetics or prediabetics when we went low carb. I strongly suspect this was true in my case.

And just exactly how was the presence of diabetes determined? I have written Lawrence de Koning, PhD to ask this question, but have yet to hear back. Was it by fasting glucose levels? Glucose tolerance test? By A1C levels? Some other method of diagnosis I'm unaware of?

Often diabetes is diagnosed by fasting blood sugar levels. As I have mentioned here before, my fasting blood sugar runs a tad high, often in the 110 range or thereabouts. But my A1C, the indicator of blood sugar levels over a few weeks time, is absolutely dead normal. I was concerned about my high fasting blood sugar until I learned an interesting fact: Low carbohydrate diets induce a degree of insulin resistance, and that insulin resistance is not abnormal nor dangerous, but healthy. Since, after a few weeks of low carbing, most of your cells are happily acclimated to running on free fatty acids and/or ketones for fuel, they don't need much glucose. So they start resisting insulin so as to shunt what glucose the low carber does have in their system to the few cells that are dependent on it.

It is for this reason that low carbers are supposed to up their carb intake to at least 150 grams per day for a few days before taking a glucose tolerance test. If they don't, the results will be skewed wildly toward high blood sugar. And this is the reason why, after fifteen years of low carbing, I have mildly elevated fasting blood glucose, even though my A1C shows that overall my blood glucose is absolutely normal: That elevated fasting glucose is normal for me.

Having said all of this, I should point out that the researchers feel that the big problem from red meat (and processed meat; let's not forget processed meat) is iron. It's well-established that iron is, indeed, one of those nutrients you can get too much of. Among other things, too much iron can cause heart disease. Wouldn't surprise me if it could mess with glucose metabolism, too.

It must be noted that animal sources of iron contain a specific form, called heme iron, which is more efficiently absorbed than the non-heme iron in plant sources. It is this heme iron that the research claims to implicate. Iron from all animal sources, including fish and poultry, is in the heme form. Yet the article suggests, It's important to replace red and processed meat with chicken and fish and also vegetable sources of protein and fat, so nuts and legumes would be top choices. Which, except for the legumes, fit nicely into our "Atkins-style program." But wait a minute: Fish? The #1 source of iron, according to the USDA, is clams.

But must we fear red meat?

A look at The USDA Nutrient Database list of food sources of iron shows that of the first fifty foods on the list, not one is an unprocessed red meat. Heck, you have to go down to number 47 to find a processed red meat product: braunschweiger, aka liverwurst. Thirty-seven of the top fifty are grain products, mostly processed cereals, but also "enriched" grain products like flour and white rice. Also included in the Top 50 are soy beans, lentils, and canned white beans. A fast-food burger comes in at number 54, but keep in mind that it's listed as "1 sandwich," which ostensibly includes the enriched-flour bun.

Again, the iron in those grain sources is in the less-assimilable non-heme form, but the doses are far greater than we're getting from meat. I had to go all the way down to number 157 to find just beef, not processed, not with a bun or a tortilla, not in a fast-food meal, and not in the form of organ meats, which most Americans reject. At 157 I found beef chuck, something I eat pretty freely.

But how much iron does that chuck have? A 3-ounce serving, with all the fat trimmed off, contains 3.13 mgs. Now, I'm likely to eat more than 3 ounces. Around here, hamburger patties -- made from ground chuck -- weigh 6 ounces before cooking. So we're looking at 6.26 mgs of iron, if the folks who ground that chuck trimmed off every scrap of fat, which I'm betting they didn't. Leaving 1/8" of fat on your chuck roast -- and I'd leave more than that -- drops the iron-per-3-ounces to 2.62 mgs. I'm doubting my 6 ounce hamburger has more than 5 mgs of iron in it. Ah, found it: Ground beef with 80% lean/20% fat -- about what I get -- has 2.11 mgs of iron per 3 ounces. So my hamburger has 4.22 mgs of iron in it.

And how much iron do we need every day? The RDA is 27 mgs. Yep, I'd have to eat more than 6 hamburger patties per day to blow past the RDA.

If I was feeling a little more flush, and went for an 8 ounce sirloin steak, I would get 3.9 mgs of iron -- again, that's if I trim all but 1/8" of the fat, which, trust me, ain't happening. How about pork? I actually eat more pork than beef. I'm a big fan of pork shoulder steaks. I eat pretty big ones, too. Let's give me a really generous meal, and figure I'm eating 12 ounces of pork shoulder. I'm all the way up to 5.48 mgs of iron. No way I'm eating 5 of those babies in a day, but that's what it would take to get the RDA of iron.

Now, that RDA is very likely set with the idea that you'll be getting much of your iron from the non-heme sources, especially all those lovely enriched cereal products. But still, I find it hard to get worked up about less than 1/6th of the RDA.

It's popular for low carb detractors to assume that we're on the all-bacon-cheeseburger-all-the-time diet, but I don't eat red meat every day, and I certainly don't eat it for every meal. I eat eggs, I eat chicken, I eat cheese, all sorts of lovely animal protein. And I'm betting most low carbers do the same.

In short, I find this whole thing a less-than-compelling mess, at least as reported in the media. Maybe there's some brilliant point I'm missing, but I doubt it.

One other thing: Though this study was presented to a meeting of the ADA, it was not sponsored by them. It was, instead, sponsored by the Canadian Institutes of Health Research and the Canadian Diabetes Association. And who sponsors the CDA? Looks remarkably similar to the ADA's list of sponsors.