Letters Section Editor: Stephen J. Lurie,
MD, PhD, Senior Editor.
In Reply: In response to Dr Hathcock, the predefined
primary outcome of our trial was coronary angiographic change, with worse
ranks assigned to women who died or had an intercurrent MI. We agree that
there is no 1-to-1 relationship between clinical outcome and angiographic
change, which is why women who died or had an MI were not imputed an equivalent
angiographic score, but rather were assigned a rank that was lower than the
rank of the worst angiographic score. The nonparametric test we used thus
relies on the relative ordering of the outcomes, not their numeric value,
and we do feel it is reasonable to rank deaths and MIs lowest.
Waters DD. Vitamins and Hormone Therapy for Coronary AtherosclerosisVitamins and Hormone Therapy for Coronary Atherosclerosis. JAMA. 2003;289(8):982. doi:10.1001/jama.289.8.982-a