Comparison of false positive rates for screening breast magnetic resonance imaging (MRI) in high risk women performed on stacked versus alternating schedules

SpringerPlus

Table 3 False positive rates according to schedule

	Overall	Stacked	Alternating	p value
	(n = 371)	(n = 165)	(n = 206)	p value
BI-RADS 3*	22 (5.9%)	16 (9.7%)	6 (2.9%)	0.006
BI-RADS 3*	[3.9, 8.9]	[6.0, 15.3]	[1.2, 6.4]
BI-RADS 4, 5, 0	29 (7.8%)	14 (8.5%)	15 (7.3%)	0.668
BI-RADS 4, 5, 0	[5.5, 11.0]	[5.0, 13.8]	[4.4, 11.8]
BI-RADS 3, 4, 5, 0	51 (13.7%)	30 (18.2%)	21 (10.2%)	0.026
BI-RADS 3, 4, 5, 0	[10.6, 17.6]	[13.0, 24.8]	[6.7, 15.1]
BI-RADS 3, 4, 5, 0	31 (24.6%)	20 (27.4%)	11 (20.8%)	0.393
For baseline MRI	[17.9, 32.8]	[18.4, 38.6]	[11.8, 33.6]
BI-RADS 3, 4, 5, 0	20 (8.2%)	10 (10.9%)	10 (6.5%)	0.230
For subsequent MRI	[5.3, 12.3]	[5.8, 19.0]	[3.5, 11.8]
Benign biopsy	22 (5.9%)	10 (6.1%)	12 (5.8%)	0.92
Benign biopsy	[3.9, 8.9]	[3.2, 10.9]	[3.3, 10.0]

*p value compares stacked to alternating schedule.
**Numbers in parentheses are percentages, and numbers in square brackets are the 95% Agresti-Coull confidence intervals for the percentage.