Although some diets may be easier for you to stick to or be more satiating, the bottom line is that you need to eat less to reduce body weight.

Africa Studio/Shutterstock

One problem in studying the effects of dietary composition is that it is not possible to vary the proportion of one component without changing the others. At the extremes of weight-loss diets, the Atkins and South Beach diets are low carbohydrate but high fat, while the Ornish diet is low fat, high carbohydrate [1]. To compare the effects of such diets outside metabolic wards, researchers must deal with study subjects whose dietary and other behaviors are not easily controlled.

Investigators do everything they can to encourage compliance with study protocols. But they confront a major challenge: Telling free-living people what you want them to do does not necessarily mean that they will follow your instructions or tell you the truth about what they are eating. And you have no easy way of getting around this problem. Because dietary intake methods all depend on accurately disclosing what subjects consume -- something impossible for most people to do -- the lack of an easy way to measure true calorie consumption in weight control studies must be considered "the fundamental flaw of obesity research [2]."

But that's not the only problem. When conducting clinical trials that compare one diet to another, researchers also face challenges in enrolling enough study subjects to satisfy statistical requirements, getting study subjects to stick to the prescribed diets, and retaining participants in the study throughout its length. Furthermore, clinical trials of diet and weight loss are expensive to conduct, and few are able to last long enough to observe whether initial weight losses were regained. These considerations make it especially difficult for investigators to evaluate the results of dietary studies objectively and for others to interpret the significance of the findings. Keep these caveats in mind as we take a look at some of the studies attempting to find out whether varying the proportions of protein, fat, and carbohydrate makes any difference to weight loss in real life.


Atwater Values indicate that fat has more than twice the energy value of either protein or carbohydrate. It makes sense to think that cutting down on fat would help with weight maintenance or loss. In the United States the various editions of the Dietary Guidelines have long promoted lower-fat diets: "Avoid too much fat" (1980, 1985), "Choose a diet low in fat" (1990, 1995), "Keep total fat intake between 20 to 35 percent of calories" (2005), and "Reduce intake of solid fats" (2010). The more recent editions have focused on limiting saturated fat and cholesterol intake rather than total fat per se in recognition of the potential role of these components in heart disease risk. But the newer guidelines also recognize that from the standpoint of body weight, calories from fat are no different from calories from any other source.

This is a shift from the earlier recommendations that reshaped the marketplace. In the early 1990s, advice to reduce fat intake was all that food companies needed to hear to start making low-fat versions of many common foods -- low-fat cheese, mayonnaise, and peanut butter, for example -- along with oxymoronic products such as fat-free half-and-half and fat-free (but equally caloric) cookies. Such products are not necessarily healthier than the products they replace, and rarely taste as good.

But the relationship of dietary fat to obesity is still of much interest. For one thing, it takes hardly any energy to store excess fatty acids as body fat, whereas it takes a bit more energy to make fatty acids from excess dietary carbohydrate. For another, proponents of low-fat diets cite experimental observations demonstrating a connection between fat intake and overweight:

  • Laboratory animals fed high-fat diets generally become obese.
  • Populations consuming low-fat diets maintain lower body weights.
  • Some clinical studies show that reducing dietary fat can result in modest weight loss [3].

Some experts, however, view such evidence as not at all specific to fat, as it could just as easily relate to high-calorie diets from any source. Low-fat diets are necessarily high in carbohydrate -- the calories have to come from something. The range of protein in diets is typically 10 percent (low) to 20 percent (high) of calories; it can't be more, because foods are low in protein -- we don't need much. The real issue in real diets is carbohydrate v. fat. Few studies of such difference control for calories. Overall, studies of dietary patterns typically find no association between either the amount or the type of fat in the diet and subsequent weight gain over periods of several years [4].

When investigators compare the effects of weight-loss diets varying in fat content, they find little difference. One study, for example, looked at overweight or obese subjects who had reduced their body weights by about 25 pounds by consuming a diet of 800 calories a day -- a reduction that ought to induce weight loss in anyone. The participants were divided into groups and instructed to consume specified diets containing 20 to 45 percent of calories from fat. Because all participants regained weight at about the same rate during the study period, the investigators concluded that the percentage of dietary fat made no difference [5].

Another study, this one of nearly 50,000 women, compared the effects of low-fat to usual diets over a six-year period. The women assigned to the low-fat diet were treated more attentively, and perhaps for this reason lost more weight during the first year. They also had lower levels of body fat. Those who best adhered to the low-fat diet kept the weight off the longest. But by the end of the study period, the difference between the two groups was too small to be statistically significant. The one long-term benefit seen in the women on the low-fat diet was a small decrease in body fat [6]. Was this benefit due to the low-fat intake or to the reduction in calories? Our guess is fewer calories.

Although diets with varying proportions of fat, carbohydrate, and protein may be easier for you to stick to or be more satiating, the bottom line is that if you want to reduce your body weight, you still need to consume fewer calories.

TEMPLATEReadMoreBookExcerpts.jpgExcerpted from Marion Nestle and Malden Nesheim's Why Calories Count: From Science to Politics (University of California Press)

[1] Atkins RC. Dr. Atkins' New Diet Revolution. New York: Avon Books, 2002. Agatston A. The South Beach Diet. Emmaus, PA: Rodale, 2003. Ornish D. Eat More, Weigh Less: Dr. Dean Ornish's Life Choice Program for Losing Weight Safely while Eating Abundantly. New York: HarperTorch, 2001.

[2] Winkler JT, The fundamental flaw of obesity research. Obesity Reviews 005;6:199-202.

[3] Bray GA, Popkin BM. Dietary fat intake does affect obesity! AJCN 1998:68:1157-73.

[4] Willett WC. Dietary fat and obesity: An unconvincing relation. AJCN 1998;68:1149-50. Forouhi NG, et al. Dietary fat intake and subsequent weight change in adults: Results from the European Prospective Investigation into Cancer and Nutrition cohorts. AJCN 2009;90:1632-41.

[5] Due A, et al. Comparison of 3 ad libitum diets for weight-loss maintenance, risk of cardiovascular disease, and diabetes: A 6-mo randomized, controlled trial. AJCN 2008;88:1232-41.

[6] Carty CL, et al. Low-fat dietary pattern and change in body-composition traits in the Women's Health Initiative Dietary Modification Trial. AJCN 2011;93:516-24. For a comment on the importance of adherence, see: Bray GA. Is dietary fat important? AJCN 2011;93:481-82.

We want to hear what you think about this article. Submit a letter to the editor or write to