From the headlines, you’d think that just eating more whole grain would enable you to live to a very ripe old age.
It turns out that things are more complicated than the headlines. There are many factors determining when you bid the world good-bye, with how much whole grain you chewed your way through possibly being one.
Or possibly not.
The study prompting these headlines was published on March 24, 2015, in BMC Medicine. It reports on the eating habits and mortality of a group of more than 350,000 Americans, most of them white. When the dietary data was collected, the subjects ranged in age from 50 to 71. The mortality statistics came from 14 years later.
The study is another in a long list of observational studies that seek to find associations between diet, death, and disease. Like most such studies, it found what it was seeking.
Odd how that happens.
As is typical of the genre, the study relied on the subjects’ self-reporting of their eating habits and lifestyles. It also relied on reporting its results in the most attention-grabbing manner: as relative risk. So, for instance, during the 14 year period covered, the study reports that the group who consumed the highest amount of whole grain had 17% less risk of dying from any cause than the group that consumed the least.
Naturally, the absolute risk rates and the difference between them are not reported. You need to figure those out for yourself from the data in the tables.
Highest whole grain intake group — 8,024 deaths out of 41,249 subjects (rate = 19.5%).
Lowest whole grain intake group — 11,845 deaths out of 41,248 subjects (rate = 28.7%).
Highest cereal fiber intake group — 7,918 deaths out of 73,489 subjects (rate = 10.8%).
Lowest cereal fiber intake group — 11,700 deaths out of 73,488 subjects (rate = 15.9%)
According to the researchers, “In specifying numbers of deaths by intake quintile, the number of deaths is determined by the energy-adjusted intake quintile for the entire population.” I believe that statement is in English, and having been a college English teacher for many years, I ought to know. I’m not entirely sure what it means, but I’ll take the numbers at face value. However, the cereal fiber data makes more sense.
So the difference in absolute rates of death between those eating the most whole grain and those eating the least is 9.2%. For the cereal fiber data, the equivalent difference is 5.1%.
A 9% absolute risk-rate difference is significant enough to warrant a closer look. Unfortunately, when you take a closer look, the problems with the study quickly become apparent. Again, as is typical in observational studies, there is no control of key variables. While they attempt to make the usual clever statistical adjustments, the researchers admit that “the positive associations may still be related to residual confounding of non-measured covariates.”
In other words, the groups being compared differed in a number of ways, not just in the amount of whole grain consumed, so it’s hard to tell what was really going on.
I’m not convinced the researchers successfully adjusted for the “covariates” they anticipated, let alone the ones they didn’t.
For starters, those eating the lowest amount of whole grain and cereal fiber were more likely to be male, obese, and sedentary than those eating the highest amount.
And those eating the lowest amount of whole grain and cereal fiber were also more likely to smoke tobacco and drink alcohol. They also consumed significantly more calories on average per day. They were less likely to have graduated from college, and were more likely to be married. (They also ate more red meat, but only 1.5 ounces more per day on average.)
It sounds like a stressed out bunch to me. Relatively, anyway. And while they weren’t eating a lot of whole grain, I suspect they were eating plenty of refined grain. Sugar, too. We know from Table 1 that those eating the lowest amount of whole grain and cereal fiber reported an average daily calorie intake that was 500 calories greater than that of those eating the highest amounts of whole grain and cereal fiber. Where did those extra calories come from? Reported fruit and vegetable intake was virtually the same for the two groups. The low whole grain/ cereal fiber group reported more red meat consumption (about 110 calories a day more on average) and more alcohol consumption (about 170 calories a day more on average). That leaves 200 plus calories a day unaccounted for. More than that, actually, since this group was by definition consuming far lower amounts of foods with whole grain and cereal fiber. All those extra calories had to come from somewhere. My guess would be from potato chips, white bread, soda pop — and various other sugary, starchy stuff.
In short, I don’t see them as low-carb, high-fat eaters.
You can be sure they weren’t following a Paleo regime.
From the evidence in the study (such as it is, given that the base data comes from self-reports), I conclude that a statistical association exists between such factors as exercise, obesity, smoking, drinking, education, marriage, maleness, and/or diet, and life span.
However, to be safe, I would like to add genetics to the list of factors.
Ah, those confounding covariates! It really is hard to tell what’s going on. Damned near impossible in a study like this. What we need are more clinical studies with controlled variables.
Still, the researchers offer an idea in their discussion that is worth pursuing. They say, “the anti-inflammatory effects of dietary fiber may help explain, at least in part, the inverse associations of whole grains and fiber consumption with chronic disease death.” So reducing chronic inflammation may be the key.
That’s interesting. Another excellent way to reduce chronic inflammation is to eat a low-carbohydrate diet. (See, for instance, Volek and Phinney, The Art and Science of Low Carbohydrate Living, p. 84-85.)
Whatever factor or combination of factors we might credit for it, the fact is that some people in the study lived longer than others. But are we talking about a miraculous old age for anyone involved?
No, we are not. When the diet and lifestyle data were collected in the mid-1990s, the mean age of the respondents was just under 62. So, 14 years later when the deaths were tallied, the mean age of the survivors was just under 76. That’s old, I guess, but not Methuselah old. Not when you consider that the overwhelming majority of the subjects were white, and that in 2007 the average life expectancies for white males and white females in the United States were 75.8 and 80.7 years respectively (CDC).
I’m not saying that some people in the study won’t live to be truly ancient, but the data stops well short of proving that. And to be fair to the original researchers, they weren’t trying to prove that.
The headlines to the contrary are just plain silly.