Fig. 3From: Pre-natal exposures and breast tissue composition: findings from a British pre-birth cohort of young women and a systematic reviewIndirect and direct effects (relative change in geometric mean) of a) maternal height and b) maternal mammographic density accounting for the mediating effect of birthweight, and birth weight accounting for the mediating effect of c) height and d) BMI at age 21 years, on MRI percent water. RC Relative percent change, BMI Body mass index. MRI-based percent water was log-transformed for the analysis, and exponentiated estimated regression parameters, with 95Â % CIs calculated by exponentiating the original 95Â % CIs, are presented. Model shown in (a) was adjusted for age, BMI and menstrual phase//hormonal contraceptive use at the time MRI, maternal education level, pre-pregnancy BMI, and smoking during pregnancy. Model shown in (b) was adjusted as for (a) plus maternal height, and maternal age and BMI at the time of mammography. Model shown in (c) was adjusted for age, BMI and menstrual phase/hormonal contraceptive use at the time of MRI, maternal education level, height, and smoking during pregnancy. Model shown in (d) was adjusted for age, BMI (linear and quadratic term), and menstrual phase/hormonal contraceptive use at the time of MRIBack to article page