Which diet can help people lose weight and keep it off? So far, no one has found a magic bullet.
“We had three decades of low-fat, and we had a decade of ‘Oh, wait, no, maybe low-carb,’ and then at the end of that we said ‘Oh, never mind, neither of them works,’ ” says Stanford University’s Christopher Gardner.
But several glimpses of new evidence are giving researchers renewed hope. They’re looking not just at how many calories people eat and burn, but at their genes, the microbes in their gut, how much they sleep, and more.
Here are some clues that may tip the scales in your favor.
Christopher Gardner is the director of nutrition studies at the Stanford Prevention Research Center and is an associate professor of medicine at Stanford University. He has done research on weight loss, vegetarian diets, soy foods, and garlic. He is a member of the Obesity Society, the American Heart Association’s Council on Nutrition, Physical Activity and Metabolism, and the scientific advisory board of the Culinary Institute of America. Gardner spoke to Nutrition Action’s Bonnie Liebman by phone from Palo Alto, California.
Q: Why are you testing low-fat vs. low-carb diets for weight loss?
A: For decades, we said “Eat a low-fat diet to lose weight,” and then the obesity epidemic kept going. So a bunch of people said “No, it has to be a low-carb diet,” though we had no data to know if that was correct or not.
So ten years later we have data. And if you pool all the studies, there’s no difference in weight loss between low-fat and low-carb diets.
But even more disappointing, neither is very good for the average person. People aren’t losing a lot of weight and keeping it off, regardless of which diet they follow. You can argue about petty differences, but when you look at average weight loss, it’s hugely disappointing.
Q: People lose only about five or ten pounds after a year?
A: Yes. But if you look at who lost a lot of weight and who lost very little weight, the range is stunning.
Q: Was that true in your trial?
A: Yes. Our A TO Z Weight Loss Study randomly assigned 311 overweight or obese premenopausal women to one of four diets. Atkins was the lowest-carb and Ornish was the lowest-fat. The Zone and the LEARN diets were in the middle.
After one year, the average weight loss was ten pounds for Atkins, six for LEARN, five for Ornish, and three and a half for The Zone.
Q: So Atkins was slightly better?
A: Yes, but no one’s going to get excited about ten pounds. For these women, who wanted to lose 15 to 100 pounds, we couldn’t say “Congrats, the study’s over. On average, you lost five or ten pounds. Everybody jump up and down.” They’d say “I didn’t even go down a dress size.”
But what about the women in the study who lost 30 or 40 pounds? They went down multiple dress sizes. And what about those who tried their hardest and lost nothing or even gained a few pounds?
Q: How much did weight loss vary?
A: In every group, some women lost 40 or 50 pounds, and some gained five or ten pounds. So we have a 60-pound range of weight change over 12 months.
We wondered what could explain that difference. We found two small feeding studies suggesting that insulin-resistant people lost more weight on a low-carb diet, and people who were not insulin resistant lost more on a low-fat diet.
Q: If you’re insulin resistant, your insulin doesn’t work properly?
A: Yes. So we went back to our A TO Z data and did a post hoc analysis, which isn’t definitive because the study wasn’t designed to look at insulin resistance. All we had were fasting insulin levels, which are a crude measure of insulin resistance.
We assumed that the third of the women with the highest fasting insulin were probably more insulin resistant, and the third with the lowest fasting insulin were probably less insulin resistant.
And we found that those who were more insulin resistant lost more weight on the low-carb Atkins diet than on the low-fat Ornish diet, which is higher in carbs. Those who weren’t insulin resistant did no better on any one diet.
Q: Is that because they stuck to their diets better?
A: Possibly. We think some people have a harder time adhering to a diet because it’s the wrong one for them metabolically.
Q: How can people tell if they’re insulin resistant?
A: It’s not easy to find out, but you’re more likely to be insulin resistant if you meet the criteria for the metabolic syndrome.
It’s exciting because it may explain part of the difference between success and failure.
If you just randomized everyone to one of several diets, the successes and the failures might cancel each other out and make it look like no diet was better than any other.
If you tease them apart, there’s a bigger difference.
Q: How much more could someone lose on the right diet?
A: We don’t know. It may be only an extra five pounds, but it could be considerably more. We won’t know until we finish our new year-long study on 600 people.
Q: Is insulin resistance common?
A: Yes. Roughly a third of the women in the A TO Z Study met the criteria for the metabolic syndrome.
So much of the country is now overweight or obese. That means that more people are insulin resistant. So if a low-fat diet—which is high in carbs—is worse for people who are insulin resistant, that diet is less appropriate now than it was before the obesity epidemic.
Q: Shouldn’t people with insulin resistance cut back on carbs anyway?
A: Yes. Cutting carbs, and especially added sugars, can help lower triglycerides, which is part of the metabolic syndrome.
Q: Does insulin resistance alone explain why people lose so little weight in most studies?
A: Not entirely. Another issue is that when we put people on low-carb or low-fat diets, we’re not very good at differentiating the diets.
If you look at the biggest and longest study—the Pounds Lost Study—it randomly assigned 800 people to four different diets for two years. The original design was elegant because the four diets had two levels of fat, two levels of protein, and four levels of carbs. But people didn’t adhere to the diets well.
Q: So in practice, the four diets weren’t that different?
A: Right. The weight-loss trajectories were identical in the four groups, and the diets eaten by all the participants were pretty similar. So why would you expect a different outcome?
In our new large study, we’re pushing people to huge extremes—Atkins, Ornish, even beyond that—but we’re anticipating and allowing them to scale it back, because people always do in these studies.
We’re asking them to go as low as they can in either fat or carbs. They haven’t been on the diets for 12 months yet, but boy, at six months we’re seeing really different diets.
Q: When will you have results?
A: Put it on your calendar to call me in 2017, and I’ll tell you what happened. It’ll be a long time before we have all 600 people on the diets for a year.