One Size Diet Does NOT Fit All

Abstract & Commentary

By Barbara A. Phillips, MD, MSPH, Professor of Medicine, University of Kentucky; Director, Sleep Disorders Center, Samaritan Hospital, Lexington. Dr. Phillips reports no financial relationship to this field of study.

Synopsis: A low-glycemic load diet is more likely to produce weight loss in obese young adults who have high insulin concentrations than is a low fat diet; it is also more likely to improve high-density lipoproteins (HDL) and triglycerides, regardless of insulin levels.

Source: Ebbeling CB, et al. Effects of a low-glycemic load vs a low-fat diet in obese young adults. A randomized trial. JAMA. 2007;297:2092-2102.

Based on the observations that high-glycemic food intake induces more insulin secretion than low-glycemic food intake and that high insulin levels may stimulate appetite, these authors hypothesized that those people whose insulin response to a glycemic load was highest might have the most trouble losing weight on a high glycemic diet. (Glycemic load, by the way, is based somewhat on the carbohydrate availability of food. In general, complex starches have lower glycemic indices than do simple sugars, and fat and protein have very low glycemic indices.) To test their hypothesis, the investigators in this study recruited (through the media, including the internet) and randomized 73 obese young (average age 28 years) adults, mostly women. They do not report the mean Body Mass Index in this paper, but the mean percentage of body fat of this cohort was about 40%. Insulin secretion was assessed using an oral glucose tolerance test; the cut-point between high and low insulin secretors was 57.5 microunits/mL after 75 grams of oral glucose. The high and low insulin secretors were randomized to either a low-glycemic (40% carbohydrate, 25% protein, 35% fat) or a low-fat (55% carbohydrate, 25% protein, 20% fat) diet. Dietitians were heavily involved in this study, conducting education and follow-up throughout the study, and assessing adherence to assigned diet. The dietary instructions themselves were general (for example, those in the low glycemic arms were to eat nonstarchy vegetables, legumes, fruits, nuts seeds and oils, and to limit sweets and starches).

Food intake was governed by simple instructions about satiety. The study lasted for 18 months, with the first 6 months being "intensive intervention," involving multiple workshops, counseling sessions, phone calls, and other educational efforts. Dietary glycemic load was calculated from food diaries. The investigators also collected data on physical activity, satisfaction with diet, and lipid levels. Because of drop-out and missing data, complete data analysis at 18 months is only available for 35 participants. For the entire cohort of 73 participants, weight loss did not differ between the two dietary groups. However, for the high-insulin secreting group, those who were randomized to the low-glycemic diet lost significantly more weight than those in the low fat diet (5.8 vs 1.2 Kg). In those who had low insulin levels after glucose challenge, there was no difference in weight loss between the 2 diets. Insulin response to glucose was not related to changes in lipid levels, but those in the low-glycemic diet had more improvement in high-density lipoproteins (HDL) and triglycerides, and those in the low-fat diet had more improvement in the low-density lipoproteins (LDL).


The lay press is full of information (and misinformation) about diets and weight loss, and often picks up on medical reports. The headline about this particular paper in US News and World Report exclaims, "Winning at losing. Your body's use of insulin may point to the right diet."1 The article goes on to assert that it's time to rethink the one-size fits all approach to weight loss, and I agree. At the end of the day, caloric balance (out vs in) determines weight loss or gain. But the composition of those calories may profoundly affect appetite regulation. The US News and World Report article goes on to recommend getting a glucose tolerance test to determine if one is a high or low insulin secretor (expecting your physician, no doubt, to justify it to your insurance company), or to "tinker with your eating plan to see what works best." Because of concerns about effects of low-glycemic (which are generally higher in fat) diets on lipid levels, many clinicians have been reluctant to recommend such diets. It is noteworthy that in the current study, those in the low-glycemic diet had more improvement in HDL and triglycerides, and those in the low-fat diet had more improvement in the LDL.

A recent randomized trial2 compared the Atkins (carbohydrate restriction), Zone (macronutrient balance), Weight Watchers (calorie restriction), and Ornish (fat restriction) diet groups. The main determinant of weight loss in this comparison was adherence to the diet, but only about half of the participants completed the trial. Notably, each diet significantly reduced the LDL/HDL cholesterol ratio. Previous work has demonstrated that low-carbohydrate (higher fat) diets can improve the metabolic syndrome,3,4 and some work (larger than the current study) has shown that low glycemic (low carbohydrate, higher fat) diets may benefit the LDL cholesterol,5,6 and reduce coronary heart disease.7

This paper helps to explain why some folks do well on a low-fat diet and others don't: those whose insulin response to a glycemic load is highest are less likely to be able to control their appetite (thus weight) in a low-fat diet. This phenotypic difference may help to explain some of the conflicting evidence about which diets "work."

Like many of our patients will, I Googled "glycemic index," and got over 1,200,000 hits. I went to the "Home of the Glycemic Index,"8 and found this information about how to switch to a "this for that" approach—ie, swapping high GI carbs for low GI carbs. You don't need to count numbers or do any sort of mental arithmetic to make sure you are eating a healthy, low GI diet.

  • Use breakfast cereals based on oats, barley and bran
  • Use breads with whole grains, stone-ground flour, sour dough
  • Reduce the amount of potatoes you eat
  • Enjoy all other types of fruit and vegetables
  • Use Basmati or Doongara rice
  • Enjoy pasta, noodles, quinoa
  • Eat plenty of salad vegetables with a vinaigrette dressing

I also found a very cogent (but much more sophisticated) explanation at Wikipedia.9 Odds are, many of our patients have been to these places already.

The bottom line is that different diets might work better for different people. After decades of vilification, fat is back!


1. Kotz D. Winning at losing. Your body's use of insulin may point to the right diet. US News and World Report. May 28, 2007, p73.

2. Dansinger ML. Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA. 2005;29:43-53.

3. Nordmann AJ, et al. Effects of low-carbohydrate vs low-fat diets on weight loss and cardiovascular risk factors: a meta-analysis of randomized controlled trials. Arch Intern Med. 2006;166:285-293.

4. McKeown NM, et al. Carbohydrate nutrition, insulin resistance, and the prevalence of the metabolic syndrome in the Framingham Offspring Cohort. Diabetes Care. 2004;27:538-546.

5. Sloth B, et al. No difference in body weight decrease between a low-glycemic-index and a high-glycemic-index diet but reduced LDL cholesterol after 10-wk ad libitum intake of the low-glycemic-index diet. Am J Clin Nutr. 2004;80:337-347.

6. Ludwig DS. The glycemic index: physiological mechanisms relating to obesity, diabetes, and cardiovascular disease. JAMA. 2002;287:2414-2423.

7. Halton TL, et al. Low-carbohydrate-diet score and the risk of coronary heart disease in women. N Engl J Med. 2006;355:1991-2002.