In earlier articles, I've described anecdotal evidence that my current diet plan works, but those anecdotes are not what drew me to start this diet. My mother, for example, has been a low-carbohydrate dieter for some time, but I was uninterested despite her apparent success. What finally got my interest this year was a study from the Stanford University School of Medicine, released in March.

When considering diet changes for weight loss, the two issues I consider are efficacy and safety. I could starve myself, for example, but that would be neither efficacious (beyond the first couple of days, anyway) nor safe. Diet pills may work, but they tend to be quite dangerous. The safest approaches — simple limited-calorie plans — tend to work slowly, but they do work. As far as low-carb dieting, I could see that it worked for my mother, though I wasn't sure it worked for most people. I knew other people on low-carb diets who didn't seem to be losing weight well. It could be, I reasoned, that my mother is simply taking in fewer calories than those other people, and so her "low-carb" diet is working for her simply because it's also a low-calorie diet. I had also read scattered articles that claimed the Atkins diet was dangerous, so while I hadn't investigated in earnest, I had questions about both the efficacy and safety of the diet, and that was that — until I saw the Stanford study.

The Stanford study lasted a year, and split a group of 300 women into four groups, each pursuing a different diet. The people on the Atkins low-carb diet lost the most weight (there's the efficacy) and also benefited the most in terms of cholesterol and blood pressure (there's the safety). Since cholesterol had been one of my primary concerns, this study told me that I should at least look further. I searched further and found a multi-university study from 2003 that also compared different diet approaches over a year. They reported very high attrition rates, and (possibly as a result) that gains at six months flattened out to almost no difference in weight loss after a full year (little or no efficacy, perhaps related to attrition), but similar health benefits (safety). A 2000 study from "California researchers" described benefits of a low-carb diet on people with diabetes, but I don't have diabetes, and the article was fuzzy on the parameters or source of the study. Still, that's another vote in support of safety. A short-term 2007 study found that the low-carb group lost "significantly" more weight during the weight-loss portion of the study, though the differences appeared to balance out once that phase of the diet was done. Again, the "bad" cholesterol level was lower for people on the low-carb diet. A 2002 study from Duke University (funded by the Atkins Institute) found that over six months, the Atkins low-carb diet was effective in weight loss, and did improve cholesterol levels, but they did not study other potential heath issues. They also did not make clear in the excerpt how rigorously controlled the study was. A 2003 study from the U.S. Department of Veteran's Affairs compared low-carb and low-fat diets and found the low-carb diet significantly more effective in spurring weight loss and also results in lowering of triglycerides, though uniquely among the studies I found, they report no difference in cholesterol levels.