Take a look at this headline from a Shape magazine online article – but I’m warning you, if you’re prone to head-bang-on-desk incidents like I am, you’d best don your helmet before continuing.
Low Carb Diet Linked to Shorter Life Expectancy
That’s the headline. Here’s the subhead:
If your healthy diet doesn’t include breads, rice, oats, and other whole grains, you may be missing out on a huge health perk, says new science.
And here’s the opening paragraph:
Swearing off carbs may mean forgoing health perks as well: People who ate more whole grains throughout their lives lived longer than those who didn’t, reports a new study in the JAMA Network Journals.
Better eat your bread and other grains, because a low-carb diet is linked to an early death. That’s the takeaway message. So obviously, the study being reported by the Shape magazine writer compared low-carb diets to diets rich in whole grains, right?
Wrong. The study wasn’t about low-carb diets at all. The headline and the opening paragraph are both complete nonsense. Hardly a week goes by when I don’t see some goof in the media misinterpret a study (often with help from the researchers), but I ignore most of those articles these days simply because they’re so common.
But this article … wow … I found myself asking the same question I often ask when politicians give speeches: Is this goofball knowingly dishonest, or just plain stupid?
So let’s put on our Science For Smart People hats and ask some questions about the study that prompted the Shape reporter (and others, no doubt) to conclude that a low-carb diet is linked to shorter life expectancy.
Q: Is this a clinical study or an observational study?
A: It’s an observational study. Actually, researchers dug data out of two ongoing observational studies. Here’s a quote from the study abstract:
We investigated 74,341 women from the Nurses’ Health Study (1984–2010) and 43,744 men from the Health Professionals Follow-Up Study (1986–2010), 2 large prospective cohort studies.
I’ve written about those studies before. The Reader’s Digest version is that they’re based on occasional food questionnaires, which are notoriously unreliable. Whenever I see a new analysis of the same old data from either one of these studies, I know it’s time to roll my eyes and walk away. Move along folks, nothing to see here. But for the sake of argument, let’s assume food questionnaires are reliable and observational studies actually tell us something useful.
Q: What was the actual difference?
A: Well, you can refer to the abstract for the details, but here’s what got the researchers and members of the media all excited:
After multivariate adjustment for potential confounders, including age, smoking, body mass index, physical activity, and modified Alternate Healthy Eating Index score, higher whole grain intake was associated with lower total and CVD mortality but not cancer mortality…. We further estimated that every serving (28 g/d) of whole grain consumption was associated with a 5% lower total morality or a 9% lower CVD mortality, whereas the same intake level was nonsignificantly associated with lower cancer mortality.
So people eating whole grains had lower mortality. Which leads to the next question …
Q: Compared to what?
A: Well, from the headline in Shape magazine online, you’d think researchers compared diets rich in whole grains to low-carb diets. But like I said before, that’s not the case. All this data shows is that people who ate more whole grains were less likely to die prematurely. So … if a person eats more whole grains, wouldn’t that mean he or she is eating less of something else? Which leads us to ask …
Q: If A is linked to B, could it be because of C?
A: That’s the $64,000 question. And the answer in this case is almost certainly yes. Whole grains are associated with better health outcomes, but that’s because people who eat whole grains usually choose them over refined grains. This study was conducted at Harvard, which trumpeted the results in the media and promoted the idea that there’s something especially health-enhancing about whole grains. Here’s a quote about the study from a Harvard press release:
“This study further endorses the current dietary guidelines that promote whole grains as one of the major healthful foods for prevention of major chronic diseases,” said Qi Sun, assistant professor in the Department of Nutrition and senior author of the study.
Wow, so it turns out the government dietary guidelines are correct! We just proved it here in our government-funded study! (The NIH funded the study, according to the same press release.) People who ate more whole grains lived longer, so that proves whole grains — in and of themselves — are good for you.
Uh-huh. But here are some quotes from a different Harvard press release, commenting on earlier data extracted from the same two observational studies:
Refining wheat creates fluffy flour that makes light, airy breads and pastries. But there’s a nutritional price to be paid for refined grains. The process strips away more than half of wheat’s B vitamins, 90 percent of the vitamin E, and virtually all of the fiber. It also makes the starch easily accessible to the body’s starch-digesting enzymes.
A growing body of research shows that returning to whole grains and other less-processed sources of carbohydrates and cutting back on refined grains improves health in myriad ways.
Eating whole instead of refined grains substantially lowers total cholesterol, low-density lipoprotein (LDL, or bad) cholesterol, triglycerides, and insulin levels. Any of these changes would be expected to reduce the risk for cardiovascular disease.
More recent findings from this study (the Nurses’ Health Studies I and II) and the Health Professionals Follow-Up Study suggest that swapping whole grains for white rice could help lower diabetes risk: Researchers found that women and men who ate the most white rice—five or more servings a week—had a 17 percent higher risk of diabetes than those who ate white rice less than one time a month.
In other words, the supposed magic of whole grains comes down to them being a somewhat better choice than refined grains that jack up blood sugar, triglycerides, insulin, etc. That tells us absolutely nothing about the health effects of whole grains vs. no grains.
The researchers noted that “replacing” one serving per day of red meat with whole grains was also associated with lower mortality. I put “replacing” in quotes because people in these studies don’t check a box that says I am now swapping one serving of red meat for one serving of whole grains in my daily diet. Those daily servings are the result of number-crunching by the researchers. Their conclusion just means that given what they consider a “serving,” people who ate one serving less of red meat and one serving more of whole grains lived longer.
As I’ve explained before, the “red meat” in these studies most often comes in the form of pizza, burritos, deli sandwiches, hot dogs, etc. – in other words, processed meats that are served with a generous helping of white flour. So when the researchers inform the media that “replacing” red meat with whole grains was associated with greater longevity, it could simply be the result of comparing people who eat pizza for dinner to people who eat chicken, vegetables and brown rice for dinner. That doesn’t tell us diddly about what would happen to your health if you swapped a steak for a plate of whole-wheat pasta.
The folks at Harvard may understand that (not that you can tell from their conflicting press releases), but the reporter from Shape magazine clearly doesn’t. She somehow managed to interpret this study as demonstrating a link between low-carb diets and an early death, even though the data doesn’t deal with low-carb diets at all.
To illustrate the depth of the stupidity, let’s take the smoking analogy I used in Science For Smart People and extend it a bit. Suppose we conduct an observational study of smokers and find that those who smoke filtered cigarettes have lower rates of lung cancer than those who smoke unfiltered cigarettes. The proper conclusion is that filtered cigarettes might be a better option than unfiltered cigarettes. It would be stupid to conclude that our study proves filtered cigarettes are good for you.
But our Shape magazine reporter took that level of stupidity a step further. To borrow a phrase from the comedy Tropic Thunder, she went full retard. Her headline is the equivalent of reading a press release about our observational study on smoking and then writing a headline like this:
Non-Smoking Linked To Higher Cancer Rate
Like I said, I can’t tell if she’s being intentionally dishonest or is just plain stupid. Either way, it’s not comforting to know she writes for a major health and fitness magazine.
You may now bang your head on your desk.