by Marion Nestle
Jun 27 2012

Does where calories come from matter to weight maintenance? A new study says yes, but I’m skeptical.

As the co-author of a recent book called Why Calories Count: From Science to Politics, I am well aware of how difficult it is to lose weight.

The problem

  • When you are dieting and losing weight, you require fewer calories to maintain and move your smaller body, and your metabolism and muscle activity—and, therefore, your total energy expenditure–slow down.
  • To maintain the weight loss, you need to eat less than you did before you began dieting.

But what would happen if you could adjust your diet to keep your energy expenditure from slowing down?

Enter Ebbeling et al in JAMA, with a comprehensive study to address precisely that question.  The results of the study and editorial comments on the findings demonstrate how complicated and difficult it is to obtain definitive answers to questions about diet composition and calorie balance.

  • The investigators asked whether calorie-controlled diets containing varying amounts of carbohydrate, fat, and protein, and varying in glycemic load (a measure of rapidly absorbable carbohydrates in foods) affected total energy expenditure in obese people who had just lost 10% to 15% of their weight, but were still obese.
  • They found that the diet lowest in carbohydrate did not slow down energy expenditure as much as did the low-glycemic index diet, or the one lowest in fat.
  • They concluded: “The results of our study challenge the notion that a calorie is a calorie from a metabolic perspective.”

This study took years and involved a very large number of state-of-the-art physiological measurements.

But I want to focus on the question of whether calories from all sources are metabolically equivalent.   Here’s my understanding of the study.

The methods

Ebbeling et al started by offering $2500 to obese volunteers to participate in a 7-month weight-loss trial.   In my view, the 21 subjects who finished the study worked hard for that money.

They had to participate sequentially in a:

  • Weight-monitoring phase for 4 weeks, during which they ate their typical diets while the investigators monitored their weight.
  • Weight-loss phase for 12 weeks, during which they were fed pre-prepared diets calculated to contain about 60% of their usual calorie intake so they would lose about 2 pounds a week.  The average weight loss over 12 weeks was an impressive 14.3 kg (31.5 pounds).
  • Weight-stabilization phase for 4 weeks, during which they were fed pre-prepared diets that provided the reduced number of calories needed to maintain their newly reduced weights.
  • Testing phase of 4 weeks on each of three pre-prepared test diets (total: 12 weeks).  All three test diets provided the number of calories needed to maintain the reduced weight.  During each of the 3 testing periods, investigators measured—not estimated—the subjects’ total daily energy expenditure (resting metabolism plus activity).

The composition of the diets

DIETS CARB% FAT% PROTEIN% DECREASE IN TOTAL ENERGY EXPENDITURE, Calories/Day
Weight-loss 45 30 25  Not reported
Low-fat (high-carb) 60 20 20 ~400
Low-Glycemic Index 40 40 20 ~300
Very low carb (high-fat) 10 60 30 ~100

Note that whenever one component of a diet gets changed, the other two components change too.   Because protein usually occurs in foods in relatively low amounts, a low-fat diet is necessarily a high-carbohydrate diet, and vice versa.

The results

  • The weight-loss part of this study showed that when overweight people were allowed to eat only calorie-controlled pre-prepared diets, they lost weight quickly and maintained the weight loss.
  • The test-diet part of the study showed that the diet lowest in carbohydrate (and, therefore, highest in fat) had the least effect in slowing down total energy expenditure.   The diet that slowed down overall energy expenditure the most was the one lowest in fat.

If these results are correct, people eating high-fat, low-carbohydrate diets are likely to have the easiest time maintaining weight loss.  In contrast, people on low-fat, high-carbohydrate diets are likely to have a harder time maintaining weight loss.

But does this mean that calories from different sources have different effects on metabolism?  Proponents of the Atkins (high-fat, low carb) diet say yes, according to an account in USA Today (in which I am also quoted).

I’m still skeptical.  The subjects in this study lost and maintained weight under highly controlled, calorie-restricted conditions, in which the calories came from a relatively low-fat, moderate-carbohydrate, high-protein diet (average diets contain 10% to 15% protein).

The accompanying editorial notes that heat losses are greater for protein than for carbohydrate or fat, and also raises questions about whether physical activity declined more with the low-fat (high-carb) diet than the others.  It also notes:

Each diet was consumed for only 4 weeks. A weight stabilization protocol…may not have adequately accounted for changing energy needs associated with readjustment to new diets.

These provocative results…emphasize the current incomplete knowledge base regarding the importance of dietary macronutrients and energy expenditure, especially after weight loss.

Under the relatively short, highly controlled feeding conditions of this study, the composition of the diet may indeed matter to metabolism.  But does diet composition matter for weight maintenance in the real world?

Other longer term studies of “free-living” people out and about in their communities show little difference in weight loss or maintenance between one kind of diet and another.

More research needed!

The bottom line

  • If you want to lose weight, eat less (it worked well for the subjects in this study).
  • It may help to avoid excessive consumption of sugars and easily absorbed carbohydrates.
  • Once you’ve lost weight, adjust your calorie intake to maintain the weight loss.
  • And understand that science has no easy answers to the weight-loss problem.