Christopher Gardner's research "supporting" the Atkin's Diet

Christopher Gardner is a personal friend of mine who a is Professor in Preventative Medicine at Stanford. He and I give lectures in each other's classes, and he is providing feedback on my current nutrition interventions with athletes (I have more experience doing research will cells, so his experience is very helpful). He just published a paper with his results from a one-year study where he randomly assigned 311 pre-menopausal women to one of four diets: Atkin's (high protein), the Zone (not as high protein), Ornish (high vegetables) or LEARN (the government sponsored "prudent diet" that is relatively high in carbs and low in fat; the acronym stands for Lifestyle, Exercise, Attitudes, Relationships and Nutrition). I spoke to Christopher about this study before it came out in publication because I was lecturing on the topic of different diets in my medical school course at Stanford a month ago. He told me that this study showed a marginal (but statistically significant) benefit in weight loss for the Atkin's diet over the others but that he believed this was because the Atkin's diet was so much easier to follow. It's guidelines were essentially 'just cut out the carbs and drink water.' This, as opposed to ratios of macronutrients (Zone, LEARN) or eating plants all day (Ornish). The interesting thing about this study is that it looks at BOTH the diets themselves as well as how easily they are followed by subjects for one year. The results present us with some very practical information, but this does NOT mean that the Atkin's diet is the most effective compared to other diets if and when all the diets are followed properly. For example, the American Heart Association showed in a meta-analysis that high-protein diets increase the risk of several diseases [1956-2000 Meta-Analysis, Circulation, 104 2001 1869] and a recent study comparing high-protein to high carbohydrate using either rapidly-digesting (refined) or slow-digesting (whole grain) carbs showed that the slow-digesting carb diet dropped the body fat just as quickly as high-protein but that only the slow digesting carb diet lowered total cholesterol and LDL, whereas high-protein and the rapidly-digesting carb diets both increased total cholesterol and LDL in 12 weeks [McMillan-Price J et al. Arch Intern Med 166 (2006) 1466]. The answer is that when a diet is actually followed, higher amounts of unprocessed plants (fruit, whole grains, but primarily vegetables) reduces all the fat stores in the body (sub-cutaneous, visceral AND organ fat i.e. fat in the arteries) whereas high protein is more likely to increase organ fat even if it reduces body fat overall. Again, this assumes the diets are strictly adhered to. If the diets are not adhered to very well, the diet with the simplest instructions, and the one that includes the most foods we have been wishing we could eat more of (bacon, cheese, etc) wins. When viewing the research literature over-all, including Christopher's new paper in the Journal of the American Medical Association, the picture that emerges is that refined carbohydrates, saturated fats, and excess calories in isolation and together lead to the vast majority of overweight in America AND that what people need to lose body fat is a program that is as absolutely simple as possible or it won't work as well for them. That is very sad news for Dr. Dean Ornish, whose dietary approach is extremely healthy (although in my opinion it is a bit short on fats and protein in general). A final interesting note to be made here is that Christopher showed that the Atkins subjects gained the most weight back in the second half of the one-year study. There were two previous one-year studies comparing Atkins to a "prudent" diet (one of them used LEARN) low in fat, and in each case the Atkins diet resulted in greater weight loss, but after 6 months the Atkins subjects re-gained more of their weight so that in the end the weight loss was no longer significantly different between the groups [Stern L et al., Ann Int Med 140 2004 778 and a study that compared Atkins to LEARN: Stern L et al., Ann Int Med 140 2004 778]. For a link to the Stanford Medicine article on Christopher's paper click here. For a link to a series of great quotes on this paper from many researchers and physicians (including Christopher Gardner and Dean Ornish) click here.