Preventive Medicine: Carbs, calories, and clickbait
A fascinating and well-run randomized trial of 164 adults just published in the BMJ indicating that eating a low-carbohydrate diet for maintenance of weight loss can increase energy expenditure has predictably generated widespread media attention. Some of the highest-profile entries in that barrage of brief attention, notably coverage in The Chicago Tribune and in The New York Times, massively misrepresent the study findings.
Both of these articles about the study play to, and propagate, the popular fantasy that when the right magical formula comes along, weight loss can be uncoupled from calories. But that contention, made in the media coverage of the study, but not in the well-written study paper itself, is completely at odds with how the study was done.
The authors enrolled overweight adults at the start, and put them all on the same weight loss diet. How did they get people to lose weight? By restricting calories, as they state clearly and explicitly in their own words: “During the run-in phase, energy intake was restricted to promote 12 percent (within 2 percent ) weight loss over 9-10 weeks.”
The researchers again turned to calories when it was necessary to adjust weight during the intervention. In words copied directly from the research paper: “We randomly assigned participants who achieved the target weight loss to high, moderate, or low carbohydrate test diets for a 20 week test phase. During the test phase, participants’ energy intake was adjusted periodically to maintain weight loss within 2 kg of the level achieved before randomization.”
Perhaps, in light of all the silly noise being made about calories, that last bit bears repeating one more time: “… participants’ energy intake was adjusted periodically…” In other words, even in this study looking explicitly at the metabolic effects of carbohydrate, when it was time to induce weight loss, or make sure it was maintained, the researchers turned to calories both times.
The signature finding of this study is that energy expenditure — by means as yet unknown — increases as carbohydrate intake decreases. That is not a refutation of the importance and relevance of calories and energy balance, but rather a reaffirmation of it. Calories count, and if you want to lose weight and keep it off — you need to burn first more than, and then as many as, you take in daily. If a lower carbohydrate diet helps to increase calories “out,” it would allow for weight loss and maintenance, at higher levels of calories “in.” An increase in energy expenditure is not, therefore, the “prize.” The prize would be an easier way to lose weight, and keep it off.
A vast amount of information about weight, in populations and individuals, in clinical trials and out in the world, shows no such effect however. Millions upon millions of Americans have gone on every conceivable kind of diet over recent decades, including many versions of a low-carb diet since the 1970s at least; but weight loss maintenance is the exception not rule.
Despite the sequential popularity of many branded low-carb diets (e.g., Atkins, South Beach, etc.), obesity prevalence for the overall population only continues to worsen. In RCTs that compared diet types, those assigned to low carb gained back weight just as readily as those in all other diet assignments.
I think this new paper in the BMJ is of very high quality, and very intriguing. Might macronutrient shifts, in the context of wholesome diets, shift energy expenditure? Might there be more than one to achieve this? Do such effects generalize, or only pertain to a select few? Are such effects maintained, or blunted over time by compensatory mechanisms? Important, worthy questions, all.
But it was calories these very researchers relied on when they needed their study participants to lose weight. Calories do count, even if counting them is not the best way to control how many you consume; for that, focus on eating high quality foods. Macronutrient thresholds tell us little about diet quality, and diet quality is what matters most to health outcomes over a lifetime.
Pepperoni isn’t good for you. Doughnuts aren’t good for you either. But maybe clickbait commentary and diet-study-headline-whiplash is worse for you still.