Using mammographic density to predict breast cancer risk: dense area or percentage dense area

Breast Cancer Research

Table 2 Associations between each continuously distributed mammographic measure and breast cancer

	OR^a Estimates (95% Confidence Interval)
Measurement	DA^b	PDA^c	NDA^d	ΔΧ2e	P value	BIC^f
Base model (no predictors)	-	-	-	-		1663.78
Dense area (DA)	1.42 (1.29-1.56)	-	-	53.22	< 0.001	1618.39
Percent dense area (PDA)	-	1.37 (1.25-1.50)	-	44.42	< 0.001	1627.19
Nondense area (NDA)	-	-	0.89 (0.81-0.98)	5.96	0.01	1665.65
DA + NDA	1.42 (1.28-1.57)	-	1.00 (0.90-1.10)	0.01^g	0.9^g	1626.22
DA + PDA	1.31 (1.11-1.55)	1.10 (0.93-1.29)	-	1.18^g	0.3^g	1625.04
PDA + NDA	-	1.62 (1.41-1.85)	1.27 (1.11-1.46)	11.61^h	< 0.001^h	1623.41

^aOdds ratio; ^bOR per standard deviation (2.49); ^cOR per standard deviation (17.91); ^dper standard deviation (2.80); ^e Change in -2log likelihood (relative to base model and P values for model comparisons; ^fBayesian Information Criteria; ^gCompared to dense area only model; ^hCompared to percentage dense area only model.

ISSN: 1465-542X