Table 4.

Effect estimates for mammographic density SNP, rs10995190, on automated measures of mammographic density (n = 3,552, the controls dataset)

Without adjustment for the other outcomeWith adjustment for the other outcome
OutcomeEstimate (95% CI)PEstimate (95% CI)P
A
Adjustment (i)
 VPD (raw)−0.061(−0.087 to −0.035)4.1 × 10−6−0.031(−0.054 to −0.007)0.0103
 MIP (processed)0.176 (0.111–0.240)8.5 × 10−80.111 (0.053–0.170)1.8 × 10−4
Adjustment (ii)
 VPD (raw)−0.044 (−0.063 to −0.025)8.7 × 10−6−0.032 (−0.051 to −0.014)7.8 × 10−4
 MIP (processed)0.148 (0.090–0.206)6.3 × 10−70.117 (0.060–0.173)5.4 × 10−5
B
Adjustment (i)
 APD (processed)−0.193 (−0.256 to −0.129)3.2 × 10−9−0.130 (−0.190 to −0.070)2.0 × 10−5
 MIP (processed)0.176 (0.111–0.240)8.5 × 10−80.106 (0.046–0.166)5.8 × 10−4
Adjustment (ii)
 APD (processed)−0.151 (−0.200 to −0.102)1.4 × 10−9−0.131 (−0.179 to −0.082)1.2 × 10−7
 MIP (processed)0.148 (0.090–0.206)6.3 × 10−70.118 (0.061–0.176)5.7 × 10−5

NOTE: Adjustment (i), adjusted for age, BMI, menopausal status, HRT use, parity, age at first birth, and machine types. Adjustment (ii), adjusted for age, BMI, menopausal status, HRT use, parity, age at first birth, machine types, and acquisition parameters. Point estimates, interval estimates, and P values (Wald tests) are based on estimated coefficients for the SNP in linear regression models with A, MIP and VPD as outcomes; and B, MIP and APD as outcomes.