Comparison of results from the randomized and observational designs for cancer.

a Results from the observational design analyses are compared to the randomized design analysis, and are considered to agree when the difference between hazard ratios is ≤0.15.b adjusted for the variables in Table 5.c adjusted for a propensity variable derived from a stepwise logistic regression model that selected 52 of 478 baseline variables.d users of personal calcium and vitamin D supplements were matched to non-users based on the propensity variablee multivariable-adjusted results from the Women’s Health Initiative Observational Study (WHI OS) involving 46,892 women with average duration of follow-up 7.2y.Comparison of results from the randomized and observational designs for cancer.