In a recent article by Diabetes blogger Jenny (I’ve added her to my blogroll), she covers several miscellaneous studies on nutrition and drugs. She started off the post referenced in the link saying that she didn’t think any of these studies was worth a separate blog post, but I disagree. There were at least two that I think would merit an in-depth examination, namely #5 and #6 (I left a comment on her article, which she was kind enough to answer). In this article, I’m going to comment on item #5.
Yet another study finds that A1c is a much better predictor of heart attack than a diabetes diagnosis or other “many other established risk factors” [i.e. cholesterol. (link no longer available) “Every 1% increment [over 5.4%] independently predicts a 19% higher odds of MI [myocardial infarction, i.e. heart attack] after accounting for other MI risk factors including diabetes.”
HbA1c is a measure of hemoglobin glycation. It is roughly an average of your blood glucose levels over the last few months, although there are some things other than blood glucose levels that can affect HbA1c, one of which is fructose, which I mentioned in a previous post. This is not really a study, but a “meta-study” of an observational study, which means that the researchers didn’t actually collect new data, but went on a data-mining trip through one or more other studies. In this case, the base study was the INTERHEART study.
Note that observational studies are inherently less reliable than controlled, double-blind studies. Observational studies are at best used as a guide for further research, and not for drawing conclusions. In fact, the hallmark of an observational study is that it always reaches whatever conclusion the funding source requires. I have never encountered an exception to this, including this one.
The body of this study, typical of nutritional studies in general, is a very difficult read, full of grammatical errors, very vague references, incomplete sentences, and confusing use (probably deliberate) of statistics. In my admittedly cynical opinion, none of this is accidental, but intended to make your eyes glaze over (especially if you an MD) when trying to read it — that’s so you will simply accept the synopsis as an accurate summary of the paper, and don’t try to actually understand the data. But it is clear that the lower you keep your HbA1c, the lower your risk of heart disease, and it is an independent factor that is much more important than most of the “risk factors” currently used by most MDs, including things like cholesterol (which merits a separate blog post).
And betraying the bias of the researchers, they recommend “therapies” for reducing HbA1c. They aren’t specific about what therapies to use, but they did not mention any dietary recommendations, which leads one to the conclusion that by “therapies” they actually mean drugs.
So, how do you lower your HbA1c? There are two main ways:
- The pill-pushing solution: Take prescriptions that boost your insulin production or increase insulin sensitivity, or take insulin to control blood sugar levels. This is not a complete solution, since there are other causes of glycation that don’t show up as blood sugar levels, and many of the currently-used diabetes medications actually accelerate the progression of diabetes.
- The diet solution: Reduce or eliminate fructose, trans fats (heated or partially-hydrogenated vegetable oils), and gluten intake. Keep your total carbohydrate intake under about 100g. Avoid pre-packaged and junk foods (yes, even the “certified gluten-free” and “low-carb” stuff), and eat fresh veggies, meats, fish, eggs, and (limited) fruits. This seems to be hard for a lot of people, and it is not yet mainstream, but it will usually work much better than the pill-pushing solution, and if you haven’t already killed off your pancreas completely, it’s possible that this approach will actually cure type II diabetes.
If you haven’t gone over to Jenny’s blog, I recommend that you do so, especially if you are diabetic. She shares a lot of good information there.