Table 3.

Colorectal cancer hazard ratio estimates for CEE and CEE/MPA from combined analysis of WHI hormone therapy trial and observational study data

Years from hormone therapy initiationWithout nonadherence censoring
With nonadherence censoring
CEE
CEE/MPA
CEE
CEE/MPA
HR* (95% CI)HR* (95% CI)HR* (95% CI)HR* (95% CI)
No prior hormone therapy
<21.03 (0.43-2.47)0.90 (0.44-1.83)1.10 (0.45-2.67)0.89 (0.39-2.04)
2-51.20 (0.47-3.06)0.62 (0.35-1.09)1.45 (0.42-5.06)0.73 (0.35-1.54)
>50.98 (0.47-2.03)0.62 (0.27-1.46)1.83 (0.51-6.50)0.65 (0.22-1.88)
Prior hormone therapy
<20.78 (0.27-2.21)0.54 (0.13-2.19)0.91 (0.31-2.68)0.60 (0.15-2.47)
2-50.81 (0.35-1.90)0.32 (0.09-1.15)0.42 (0.11-1.57)0.31 (0.07-1.46)
>52.46 (1.12-5.38)0.60 (0.19-1.86)3.98 (1.06-4.86)0.84 (0.20-3.55)
Ratio of HR in OS to HR in CT0.63 (0.30-1.34)1.81 (0.82-4.00)0.35 (0.10-1.23)1.62 (0.61-4.33)
  • Abbreviations: OS, observational study; CT, clinical trial.

  • * From Cox regression analyses with stratification and adjustment variables as in Table 2 footnotes, based on combined clinical trial and observational study analyses that include an interaction between hormone therapy HR and cohort (CT vs OS) that produces the ratio of hormone therapy HR in the OS to that in the CT shown at the bottom of the Table.