Table 5. Risk of breast cancer associated with IGF1 (CA)_n microsatellite genotype by racial/ethnic groupa.

Race	Genotype	CA (%)	CO (%)	OR 95%CLb	OR 95%CLc	P-trend
African American	Non-19/non-19	24 (30)	39 (43)	1.00	1.00
	Non-19/19	46 (57)	39 (43)	1.92 (0.99, 3.72)	1.99 (1.00, 3.96)
	19/19	11 (14)	13 (14)	1.38 (0.53, 3.56)	1.32 (0.50, 3.47)	0.25
Japanese American	Non-19/non-19	27 (36)	32 (34)	1.00	1.00
	Non-19/19	35 (46)	51 (54)	0.81 (0.42, 1.59)	0.81 (0.41, 1.63)
	19/19	14 (18)	11 (12)	1.51 (0.59, 3.87)	1.46 (0.56, 3.82)	0.65
Non-Latino White	Non-19/non-19	12 (15)	8 (9)	1.00	1.00
	Non-19/19	26 (32)	47 (51)	0.37 (0.13, 1.02)	0.35 (0.13, 0.99)
	19/19	44 (54)	37 (40)	0.79 (0.29, 2.15)	0.82 (0.30, 2.24)	0.59
Latino American	Non-19/non-19	7 (9)	16 (17)	1.00	1.00
	Non-19/19	46 (57)	43 (45)	2.44 (0.92, 6.52)	2.27 (0.84, 6.16)
	19/19	28 (35)	37 (39)	1.73 (0.63, 4.77)	1.76 (0.62, 4.97)	0.40
All races combined	Non-19/non-19	70 (22)	95 (26)	1.00	1.00
	Non-19/19	153 (48)	180 (48)	1.15 (0.79, 1.68)	1.14 (0.78, 1.68)d
	19/19	97 (30)	98 (26)	1.34 (0.89, 2.04)	1.40 (0.90, 2.20)d	0.13d

^aOf the 373 controls selected for this case–control data set, 134 overlapped with controls selected for the plasma IGF1 by racial/ethnic group analysis (Tables 1, 2 and 3).

^bCrude.

^cAdjusted for age as a categorical variable.

^dAdditionally adjusted for race.