Table 2.

Colorectal expression of p21, MIB-1, and hTERT during the clinical trial

Baseline6-mo follow-upAbsolute Rx effect*Relative effect95%CIPinteraction§
A. p21 expression in colorectal crypts
    Entire crypts (LI)
        Vitamin D220.810.170.08211.580.180.02200.980.340.012.421.304.51
        Ca + vit. D231.120.170.62211.130.180.60210.230.330.471.250.692.260.01
    Upper 40% of crypts (LI40)
        Vitamin D220.770.150.13211.540.160.01200.960.310.002.441.314.53
        Ca + vit. D231.020.150.70211.090.160.43210.260.310.411.290.712.340.004
B. MIB-1 expression in colorectal crypts
    Entire crypts (LI)
        Vitamin D220.830.100.18221.080.100.58220.
        Ca + vit. D231.250.090.09211.100.100.4921−
    Ratio of upper 40% to entire crypts (ϕh)
        Vitamin D220.080.010.56220.070.010.7122−0.0030.030.890.970.481.94
        Ca + vit. D230.080.010.72210.070.010.8421−0.0030.030.920.970.481.970.68
C. hTERT expression in colorectal crypts
    Entire crypts (LI)
        Vitamin D220.830.100.08220.970.100.85210.
        Ca + vit. D221.080.100.98211.060.100.70200.140.210.801.050.721.540.33
    Ratio of upper 40% to entire crypts (ϕh)
        Vitamin D220.370.010.81220.410.010.6321−
        Ca + vit. D220.390.010.25210.370.010.0320−

Abbreviations: Rx, treatment; 95% CI, confidence interval; Ca + vit. D, calcium + vitamin D.

  • * Absolute treatment effect = ([treatment group follow-up - treatment group baseline] - [placebo group follow-up - placebo group baseline]); actual calculations from the Mixed model, in which the interactions between treatment group and follow-up visits terms estimate absolute treatment effect in each active treatment group relative to the placebo.

  • Relative effect = [(treatment group follow-up/treatment group baseline)/(placebo follow-up/placebo baseline)]; interpretation similar to that for an odds ratio (e.g., a relative effect of 1.8 would indicate a proportional increase of 80% in the treatment group relative to that in the placebo group).

  • 95% confidence interval for relative effect calculated by the Delta method (24).

  • §P value for interaction between calcium and vitamin D3 treatments from Mixed model; covariates included random intercept, two factors (calcium and vitamin D3 coded as 0/1 variable), follow-up visit, and all appropriate interaction terms between factors and follow-up visit.

  • P value for difference between each active treatment group and placebo group from Mixed model. Covariates included random intercept, follow-up visit, treatment group, and treatment group by follow-up visit interaction.

  • Biomarkers detected immunohistochemically and then their labeling densities were quantified by image analysis; all biomarkers values shown as batch-standardized optical densities. Batch standardization for each biomarker was done by dividing each individual measurement by the staining batch's average optical density.