It works to lose weight, but the fundamental questions of long-term risks to health haven't been answered.
I read the NEJM article too but the study size was too damn small (132) with a HUGE dropout rate and included only the morbidly obese (BMI >40). They looked at glycemic control, lipid profiles, weight loss, and blood pressure.
The significant differences (low-carb vs. low-fat, respectively)found were:
1. Triglycerides declined at average 38 vs 7 mg/dl.
2. Insulin sensitivity (in the non-diabetic patients only) declined by an index value of 0.02 vs increase of 0.01. I think this outcome should be taken with extra grains of salt.
3. Glucose levels (I think this must be random glucose levels) declined by 26 vs 5 mg/dl. This is not bolstered by a significantly different decline in HbA1c levels.
4. Insulin levels were as expected lower in low-carb dieters.
All the rest of the factors are more or less the same. Again, I wish more studies of a wider range of patients could be conducted. I think that the inclusion of other types of diets (e.g., calorie-restricted but not calorie-type restricted) should be included. Larger study size, more diverse study population, and longer study times are needed to determine long-term outcomes (e.g. Colon cancer risk, glomerular filtration rate trends). Again, good randomization and stratification models need to be applied. I sit on my hands and wait.