Low-Carb May be Better at Maintaining a Weight Loss

Low-Carb Science in Action

plates of Mediterranean food
A Mediterranean table, with a variety of healthy foods. Cultura/Brett Stevens/Getty Image

Our bodies have a lot of tricks to encourage us to regain weight after a loss. We get hungrier as hunger hormones increase. Our brains have a more emotional response to food. Our metabolisms start to slow down. And we naturally start to move less. When I say that "we move less", I'm not just talking about what we think of as "exercise", I'm talking about the amount we move during the day just in the course of our lives (what is sometimes called NEAT).

People who have lost weight just don't feel like moving around as much. They sit more. They fidget less. It's the body's natural way of conserving calories.

Researchers at Boston Children's Hospital had a question: does what people eat make a difference in how many calories they burn up after weight loss? They developed a very well-controlled study, published in 2012 in the Journal of the American Medical Association to answer this question. They had 21 people lose weight on a calorie-controlled diet that was 45% carbohydrate, 30% fat, and 25% protein. Then they had them maintain their weight for three months, alternating for one month each on three different diets. This is called a crossover study, and has powerful advantages. Perhaps the most important one is that since each person is compared on each of the different diets, extraneous variables are minimized. Everything from genetics to the amount they sleep are more likely to be fairly constant in the three diet conditions than if they were three different groups of people.

Other strengths of the study: They gave people the food they were to eat. The number of calories they ate was customized to each person so that they would maintain their weight in all three diet conditions. And they helped to assure that everyone did what they were supposed to do by close monitoring, and paying them $2500 to complete the study (plus free food for 6 months).

The three diets were:

Low-Fat - Basically the standard diet that people are often advised to use to lose weight. Lots of whole grains (no refined grains or added sugars - none of the diets included "junk food"), vegetables, fruits, lean protein, etc. 20% fat, 20 % protein, 60% carbs

Very Low-Carbohydrate - Similar to an Atkins diet - 10% carbs, 30% protein, 60% fat. 50 grams of total carb (But here's an odd thing: this food only provided 8 grams of fiber. Although they also ate 9 grams of fiber from Metamucil, so I'm thinking that they had 33 grams of net carb from vegetables, nuts, etc. Still, that fiber figure seems low.)

Low-Glycemic Index- Of course, a low-carb diet is automatically low-glycemic -- that's the whole point of it. This diet low-GI diet was basically a moderate-carb diet. 40% carb, 40& fat, 20% carbs, with 205 grams of carb. What is the Glycemic Index?

Results: First I'm going to talk about averages, with the realization that averages do not tell the tale for any one person. But we'll get to that later. Most of the results were on a continuum, with low-glycemic being between low-fat and low-carb.

The main variable they were interested in was: how much energy was expended on each diet, both at rest (metabolism) and total?

It turns out there was a very big difference! The low-fat diet was associated with the largest reduction in metabolism, and when on the low-carb diet, people expended on average more than three hundred calories more each day than on the low-fat diet! That is, frankly, a shockingly large difference. Also, on the low-carb diet (compared to low-fat) triglycerides were 50 points lower, HDL ("good") cholesterol was 8 points higher, and insulin sensitivity was hugely improved.  Remember, this is in the same people.

Why would this be? The researchers actually don't know. They don't think it was the so-called "thermic effect of food", which some people think is a reason why people lose weight on low-carb (I don't happen to see a lot of evidence for more than a minor thermic effect).

The amount of exercise was the same in all three conditions. They ruled out other things. In an article in USA TODAY, the lead researcher, David Ludwig, is reported as saying, "we think the low-carb and low-glycemic index diets, by not causing the surge and crash in blood sugar, don't trigger the starvation response. When the body thinks it's starving, it turns down metabolism to conserve energy."

One of the things I love about the report of the study is that they show how each individual did on each diet, in terms of metabolism and energy expenditure. As we are seeing in other diet studies, there is a LOT Of individual variation in response.

Example: Although on average people expended more energy eating low-carb, two people actually expended more than 500 calories more on the low-glycemic diet than they did on either of the other two diets. Some of the people appeared to do similarly on all three diets. Also (just eyeballing the graph) the low-carb column appears to have more variation in it than the low-fat column. What does this mean? Big shock: we are not the same. (This is why I don't tell people how much carbohydrate to eat, no matter how much they beg.)

One Man's Experience: There was an interesting interview in the Los Angeles Times with one of the participants in the study. A couple of quotes from that article:

"The low-fat diet was one I would never want to do again," he said, remembering the gnawing hunger he felt after a meal that included a heaping portion of mashed potatoes with very little salt and no butter. "You would think that would be a ton of food, but I was hungry an hour later."...Rogers wasn't hungry on the low-carb diet, but he "had dreams about pizza and pasta." He felt best on the low-glycemic index diet, he said.

Remember, he was eating the exact same number of calories on each diet. What's this about all calories being the same??? They clearly do not act the same in our bodies! This is why some of us get so frustrated at the "just cut back on what you eat and it will all be fine" message.

Interestingly, the conclusion of this study wasn't that a low-carb way of eating should be recommended. Even after saying, "the low-fat diet produced changes in energy expenditure and serum leptin that would predict weight gain" and "the very low-carbohydrate diet produced the greatest improvements in most metabolic syndrome components" (as well as promoting better metabolism and energy expenditure), it was the low-glycemic one that the researchers felt came out on top, and so did most of the many news stories that covered this study. Why? Three reasons:

1) Cortisol - Cortisol is a stress hormone. When the people were eating low-fat, their cortisol levels averaged 50 (range: 41-60), low-glycemic was 60 (49-73), and low-carb was 71 (58-86). Normal cortisol is anything below 100. What does this mean? I don't know. The implication is that it's a bad sign, and related to bad things, but I truly don't know at what level cortisol becomes a problem. I presume lower is better, within some limits. For me, it brings up a questions: Is an 11 point difference so bad that it overwhelms the advantages of lower triglycerides, better insulin sensitivity, higher metabolism, etc? Would the amount of cortisol moderate over time? Open questions.

2) C-Reactive Protein - Much was made of this in the media, but I truly think this one much ado about hardly anything. CRP is a measure of inflammation, and we surely do not want that. At the beginning of the study the average CRP among the participants was 1.75, which is in the "average risk" range (1-3). During each diet, the average dropped to under one, in the "low risk" range. Low-fat .78, low-glycemic .76, low-carb .87. I, personally, will gladly take a CRP of .87, especially since I can also have normal blood pressure, lower blood glucose, low triglycerides, and all the rest of what has come for me with low-carb eating. But three years later in 2015 I'm still hearing people quote this study as saying that low-carb diets "promote inflammation" even though the CRP level was "low-risk"!  Nonsense.

And now we come to the dreaded

3) People Can't Stay on Low-Carb Diets - "Not feasible for many individuals" was the exact phrase. There is no doubt that changing to a low-carb diet (as we've seen the exact amount of carbohydrate would vary according to the individual) is an adjustment. To be fair, a big adjustment. It is clear that many people are not willing to make that adjustment, even if it would be the best thing for their health. But frankly, most of the time, people aren't given much of a chance! They aren't given the kind of support they need to go against everything they hear in the media, and to make healthy choices when the wrong kinds of foods are everywhere.

I've said it before: We don't say to people with a wheat allergy, "It's just too hard to cut out wheat: go ahead and have some bread." But we think it's fine to agree that, for people whose bodies don't process sugars and starches well, that "it's just too hard to eat what would be an optimal diet for your body. So don't even bother; eat this other diet, which is, at least, an improvement." If you have to eat that "better but not best" diet, fine. Truly. But let's not quit before we start!  This study should be encouraging health care providers really drill down and figure out the best diet for them -- it can make such a big difference in their health.

Read Gary Taubes' Commentary on this Study in the New York Times


Ebbeling, CB et al. Effects of Dietary Composition on Energy Expenditure During Weight-Loss Maintenance. JAMA. 2012;307(24):2627-2634.

Continue Reading