Secondary end points were analyzed in a predefined
order using a hierarchical testing procedure to control type
I error. These end points were analyzed with the same
methodology as for the primary end point, except
lipoprotein (a) and triglycerides, which were analyzed
using a multiple imputation approach for handling of
missing values followed by robust regression. The proportion
of patients with LDL-C b70mg/dL was analyzed using a
multiple imputation approach for handling of missing
values followed by logistic regression.