Table 2.

Mammographic density and risk of breast cancer according to method of detection: unmatched analysis and radiologists' classification of density

			Categories of percent density, percentage
		Number of pairsa	<10	10 to <25	25 to <50	50 to <75	>75	P valueb
All	Case	1,112	230	272	336	178	96
	Control	1,112	362	270	290	144	46
	ORc		1	1.75	2.06	2.43	4.74	<0.0001
	(95% CI)			(1.4, 2.2)	(1.6, 2.6)	(1.8, 3.3)	(3.0, 7.4)
Screen-detected	Case	717	173	171	219	102	52
	Control	717	242	162	196	88	29
	ORc		1	1.65	1.77	1.98	3.52	<0.0001
	(95% CI)			(1.2, 2.2)	(1.3, 2.4)	(1.3, 2.9)	(2.0, 6.2)
Non-screen-detected <12 monthsd	Case	124	12	22	33	32	25
	Control	124	35	29	29	23	8
	ORc		1	2.11	3.61	5.65	17.81	<0.0001
	(95% CI)			(0.9, 5.2)	(1.5, 8.7)	(2.1, 15.3)	(4.8, 65.9)
Non-screen-detected >12 monthse	Case	262	43	79	80	42	18
	Control	262	82	79	62	30	9
	ORc		1	2.00	2.64	3.13	5.68	<0.0001
	(95% CI)			(1.2, 3.4)	(1.5, 4.6)	(1.6, 6.2)	(2.1, 15.5)

^aNine pairs were excluded from the screen or non-screen group analysis because of missing information on detection (n = 1) or the last mammogram date (n = 8). ^bP value for the Cochran-Armitage trend test. ^cAdjusted for age, body mass index, age at menarche, parity, number of live births, age at first birth, menopausal status, age at menopause, hormone replacement therapy (ever/never), breast cancer in first-degree relatives (0, 1, and 2+), study (National Breast Screening Study, Ontario Breast Screening Program, and Screening Mammography Program of British Columbia), and observation time (2 years, 2 to 4 years, and greater than 4 years). ^dCancers detected within 12 months of the last screening date. ^eCancers detected 12 months or more after the last screening date. Table reproduced from [15]. CI, confidence interval; OR, odds ratio. Republished with permission from [15].