Clifton challenges the interpretation of the findings of our study1 comparing 4 diets of varying glycemic load on weight loss and cardiovascular risk. Unfortunately, some of his statements are incorrect. He implies that a post hoc analysis of fat loss in women was not justified because we did not first demonstrate a significant interaction between fat loss and sex. In fact, the first paragraph of our results states “Sex influenced fat mass changes, with a significant interaction with diet (P = .008).”1(p1470) In women alone, the fat mass changes were highly significant (P = .007). Clifton states that the high-protein, low-GI diet (diet 4) lowered the HDL-C ratio the most. There were no significant differences among the groups in either HDL-C level (P = .82) or HDL-C ratio (P = .11). The only significant between-group differences in lipids were in total cholesterol (P = .04) and LDL-C (P = .02) levels, favoring the high-carbohydrate, low-GI diet (diet 2).
Brand-Miller J, McMillan-Price J, Petocz P. Glycemic Load and Cardiovascular Risk—Reply. Arch Intern Med. 2007;167(2):206-207. doi:10.1001/archinte.167.2.206-b