One of the more titillating medical stories to make news recently is a study in the American Journal of Clinical Nutrition indicating that calorie intake in the United States has come down, and obesity rates have not. What makes this titillating, of course, is that it seems to suggest some great new mystery of energy balance. But I think we can account for this finding without revisiting laws of thermodynamics. We can, and we should—because there is real potential danger in abdication. If we don't account for these calories, others will.
The ad gives us a can of Coke, presumably 12 ounces, providing 140 calories. We are then shown the activities we can "enjoy" to burn up those calories. I trust everyone recognizes the activities are additive—you have to do them all to burn those 140 calories. And I trust those watching reliably do the math and reach the conclusion that it's roughly 37 minutes of physical activity all together.
Coca-Cola doesn't address how long it takes to drink those 140 calories, but we all know it's a matter of seconds. Nor does it look at the energy balance situation in reverse: You could replace the calories burned in 37 minutes of moderate activity by drinking just one 12-ounce Coke! And, of course, Coca-Cola doesn't even hint at a reality where most people who drink Coke drink more than 12 ounces and more than one, and where we can't get most people up to even 20 minutes of daily physical activity. Coke's own, happy ad indicates that if you drink two of these babies a day, you need well over an hour and a quarter of moderate physical activity to burn just those calories—to say nothing of any others you happen to consume.
No, in Coke's accounting for calories, it leaves those logical considerations to us. Because, of course, the company want this to be less about Coke, and more about our couches. If only we would get off the couch more, we could (presumably) drink Coke to our heart's content!
For now, we can leave aside other considerations—such as the quality of calories and the fact that Coca-Cola provides no nutritional value—aside. Let's get back to the new study. I'm sure folks at Coca-Cola love it, because it readily invites a "we're-doing-fine-with-calories-and-not-exercising-enough" interpretation. That's tailor-made to support Big Food's preferred answer to the problem of epidemic obesity.
So we need to account for those calories, or we invite Madison Avenue to do it for us. And we can.
First, some of the answer may well be a decline in physical activity. We have recent evidence that physical activity is being jettisoned ever more routinely from schools; that we sit more hours a day than ever before, and shorten our life spans as a result; and that sedentariness may now represent the leading cause, and certainly a leading cause, of years lost from life and life lost from years around the globe. This all suggests we should, indeed, increase our activity level. It does not suggest room for adding ever more Coca-Cola calories as we do so.
Second, I'm a bit surprised to hear that obesity rates have only kept rising. Haven't we been told from just the same kinds of studies that obesity rates had plateaued? Haven't we heard that obesity rates have actually declined in locations around the country? I am perennially frustrated by our tendency to forget about every prior medical study that made news every time a medical study makes news. OK, our calorie intake may have come down a bit. But so, I thought we had been told, had obesity rates. These two things fit together rather handily.
Third, there's the fairly obvious possibility of erroneous reporting. Getting accurate dietary intake information is notoriously difficult. We all tend to underreport our caloric intake and overreport our physical activity. We also tend to underreport our weight and overreport our height. None of this is willful deceit; it's just human nature. We tend to spin everything to the positive, ourselves included, apparently.
We also know that in all survey research, participants have some tendency to tell the researchers what they want to hear. As our society has become ever more focused on epidemic obesity, it would be no great surprise if we were ever more inclined to tell, inadvertently, little white lies about our calorie intake.
Fourth, there's the fact that before the recent decline in reported calorie intake, there was a much bigger rise. The researchers tell us average food intake went up by some 314 calories per day between the early 1970s, and 2004. It has apparently gone down by some 74 calories per day since. But that means we've cut back less than a quarter of our new-age gluttony. If we're still eating more calories than we need, we won't be getting thinner any time soon—we'll just start getting heavier less fast.
Fifth, it takes time either to gain weight or lose it. If we really did only recently dial back our average calorie intake, then maybe obesity rates will follow. But we need day after day, month after month of consistently lower calorie intake before it shows up as more than a blip on the nation's scale.
And then sixth, and finally, what I think best accounts for a decline in calories, and no corresponding change in obesity rates: the bell curve. I'll explain.
We define "obesity," for better or worse, using the BMI (body mass index) and specific cut-points. A BMI from roughly 18 to 25 is "normal" weight. A BMI from 25 to 30 is "overweight." And a BMI greater than 30 is "obese."
Now imagine that the entire population does indeed reduce average daily calorie intake—and that average weights do come down a bit as a result. Won't all of this weight loss show up as a change in obesity prevalence?
Of course not! If normal-weight people become slightly leaner, there will be no change in obesity rates. If overweight people become less overweight, but stay in the overweight range, there will be no change in obesity rates. And if obese people become less obese but stay in the obese range, there will be no change in obesity rates. The only groups in the population likely to affect obesity rates are those right at the cut-points: those with BMI of just under, or just over 30. The former can gain a little weight and make obesity rates go up; the latter can lose a little weight and make obesity rates go down. Everybody else would have to gain or lose a lot of weight to affect obesity rates at all.
The data to tell us what percentage of the population has a BMI just north of 30 are available, but I couldn't get my hands on them while writing this. So, invoking the common bell curve distribution as a default, we may reasonably infer that weight is distributed across the spectrum from low to high, clustering around the mean. That suggests that only a very small percent of the population—well below 10 percent, certainly, can lose a bit of weight and affect obesity rates. The study in question, though nationally representative, is still based on samples of just thousands, and could readily be blind to such a relatively rare occurrence.
That will do for now. I acknowledge that the quality of calories matters, but so does the quantity. We cannot allow a seemingly small divergence in calorie and obesity trends to invite wild imaginings. Yes, we should be more physically active—but since it's far easier to out-eat exercise than to out-exercise all those tasty calories, we ignore the "calories in" side of the energy-balance equation at our peril.
We can account for the recent study with no great difficulty. We can certainly do so without rewriting any laws of physics. We can account for those calories, and need to do so—because the likes of Coca-Cola have already indicated how happy they would be to do it for us.
Hungry for more? Write to email@example.com with your questions, concerns, and feedback.
David L. Katz, MD, MPH, FACPM, FACP, is a specialist in internal medicine and preventive medicine, with particular expertise in nutrition, weight management, and chronic-disease prevention. He is the founding director of Yale University's Prevention Research Center, and principal inventor of the NuVal nutrition guidance system. Katz was named editor-in-chief of Childhood Obesity in 2011, and is president-elect of the American College of Lifestyle Medicine.