Does New Study Vindicate Atkins?

A recent study of about 300 middle-aged obese men, average age 51 and average BMI 31, purported to examine the impact of a ?Low-Carbohydrate? (LC) Diet, a ?Low-Fat? (LF) Diet, and a ?Mediterranean? (Med.) Diet on weight loss and cardiovascular disease (CVD) risk factors. This study received widespread media coverage in July 2008 with some reports suggesting the results of this study vindicated the claims of the late Dr. Robert Atkins. Dr Atkins? books claimed that a low-carbohydrate, high-fat diet with lots of saturated fat and cholesterol-rich animal products not only promoted weight loss but also was heart healthy. However, a closer look at the study?s design and results suggests it does little to vindicate Atkins claim that his diet was heart healthy.

This new study was conducted in Israel and was funded by the Atkins Foundation. However, the researchers stated that the foundation did not influence their study. This is rather a curious claim since the late Dr. Atkins stated in his many books that a diet high in meat, eggs, cheese, butter and other animal fats does not adversely impact blood lipids and was heart healthy as long as carbohydrate intake was severely restricted. In this study the subjects assigned to the LC Diet were counseled to ??choose vegetarian sources of fat and protein?? and yet also claimed the LC Diet ??was based on the Atkins diet.?1Anyone familiar with the Atkins diet knows he did not primarily advocate tofu and vegetable oils in place of carbohydrate-rich foods so surely the researchers knew this as well. So why did they have their subjects following a diet they claimed was based on the Atkins diet avoid lots of meat, cheese, and butter? Perhaps they realized such a diet would raise total and LDL-cholesterol levels and promote heart disease?While the authors of this study claimed no conflicts of interest it should be noted that Dr. Stampher works with Dr. Willett at Harvard Medical School and the Med. Diet used in the study was based on Willett?s version of a Mediterranean Diet.Here are the interesting points that we believe make this study?s results deceiving:? The Med. Diet and the LF Diet had the same amount of saturated fat and cholesterol? but the Med. Diet had about 50% more fiber. Willett is a big fan of olive oil but it is fiber-free.? Had the study used a real LF Diet such as advocated by Pritikin, Ornish, and others, the LF diet would have had much more fiber and far less saturated fat and cholesterol than either the Med. Diet or the "Atkins" Diet used in the study.? Indeed, the LF Diet used in this study had the same percent of fat and saturated fat as the subject's original diet and only 9% less fat (as percent of calories) than the Atkins Diet.? Weight loss was somewhat greater on the LC and Med. Diets than the LF Diet with subjects after 2 years losing 10.3, 9.7. and 6.4 pounds respectively on the 3 diets. The dietary data reported that calorie intake was reduced by 200 calories a day more on the LF diet than on the Med. Diet, which makes little sense given the somewhat greater weight loss for those on the Med. compared to the LF diet.HDL levels rose and triglyceride levels fell roughly in proportion to the loss of weight on the 3 diets, which should surprise no one.One might look at the results of this study and conclude a LC Diet or a Med. Diet is better for weight loss than a LF Diet. However, this is a bit of a stretch because research has shown that the relative impact of modest changes in the ratio of fat to protein to carbohydrate in the diet have little or no independent impact on ad libitum energy intake. However, less dietary fiber, a higher calorie density, and more caloric beverages all tend to promote increased calorie intake. Were these variables controlled? Fiber was highest in the Med. Diet and the authors said nothing about calorie density or caloric beverages. It is quite possible the differences in weight loss on the 3 diets could be explained by differences in calorie density, fiber and liquid calories. Bottom Line:This study's results on CVD risk factors are not all that surprising given the questionable design of the study. The fact is all 3 experimental diets failed to lower Total and LDL-cholesterol levels. Many studies show very-low-fat, near vegetarian diets drop LDL-cholesterol 20 to 40%. Furthermore, the lack of attention to important dietary variables like calorie density and percent of calories consumed as beverages which do impact ad libitum calorie intake, is inexcusable today in a study that purports to examine the impact of diet on body weight regulation. Indeed, the results of this study and the media reports that followed will likely further confuse people and sadly some health professionals about the likely impact of healthy low-fat diet based mostly on minimally processed whole grains, fruits and vegetables for aiding weight loss and preventing and even reversing CVD.By James J. Kenney, PhD, RD, FACNReferences:1. N Engl J Med 2008;359:229-41