Height, adiposity and body fat distribution and breast density in young women

Breast Cancer Research

Table 4 Absolute dense breast volume differences associated with difference in anthropometric and body composition measures

	Unadjusted^a				Adjusted^b				Also adjusted for childhood or young adult BMI^c
	% Diff	95% CI	Pvalue	R ^2d	% Diff	95% CI	Pvalue	R ²	% Diff	95% CI	Pvalue	R ²
Adult anthropometric measures
Height (cm)	-8.9	-22.2, 6.8	0.25	0.01	-2.5	-12.6, 8.7	0.65	0.19	-5.6	-15.5, 5.4	0.31	0.26
BMI (kg/m²)	-20.4	-29.6, -10.0	< 0.001	0.07	-20.8	-29.8, -10.7	< 0.001	0.17	-6.8	-18.9, 7.0	0.32	0.25
Waist (cm)	-23.7	-37.7, -6. 7	0.009	0.08	-24.7	-39.5, -6.3	0.011	0.20	-15.6	-32.1, 4.8	0.12	0.28
Adult DXA measures
Total fat (%)	-24.6	-33.9, -14.1	< 0.001	0.12	-23.7	-32.9, -13.1	< 0.001	0.20	-13.8	-24.1, -2.3	0.02	0.28
A:G ratio	-26.9	-35.4, -17.3	< 0.001	0.14	-26.6	-34.2, -18.2	< 0.001	0.23	-19.6	-27.1, -11.2	< 0.001	0.30
Childhood BMI
BMI z-score at 8 to 10 years old	-28.4	-37.3, -18.3	< 0.001	0.13	-30.0	-35.9, -23.5	< 0.001	0.25	-27.0	-32.9, -20.7	< 0.001	0.25

Percentage difference (%Diff) in absolute dense breast volume associated with a one-standard-deviation difference in anthropometric and dual-energy X-ray absorptiometry (DXA) body composition measures. Results are also interpretable as the excess relative ratio. To convert to the relative ratio (RRatio), use the formula: RRatio = (%Diff + 100)/100. A:G, android:gynoid fat ratio; BMI, body mass index; CI, confidence interval. ^aEstimates from six linear mixed-effects models including clinic as a random effect and anthropometric and body composition variables as fixed effects. Anthropometric and body composition variables are modeled separately without mutual adjustment. ^bEstimates from six linear mixed-effects models as described for unadjusted plus including race, education (attended college), smoking status, duration of hormone use, and parity as fixed effects. Model for height also includes weight as a fixed effect. ^cEstimates from six linear mixed-effects models as described above for adjusted plus including BMI z-score at 8 to 10 years old as a fixed effect in adult anthropometric and DXA models and young adult BMI in the childhood BMI model. ^dProportion of variance explained by model.