Long-term prognostic implications of risk factors associated with tumor size: a case study of women regularly attending screening

Table 2 Associations between risk factors and having a tumor larger than 2 cm at diagnosis, estimated by logistic regression modeling

	Odds ratio (95% CI) for having a large vs. small tumor^a
	n = 1931
Risk factor	Crude	Multiple adjusted^b
Age at diagnosis (OR per 10 years)	0.83 (0.68 to 1.01)	0.87 (0.71 to 1.05)
BMI ≥25 vs. BMI <25	1.34 (1.08 to 1.66)	1.47 (1.18 to 1.84)
PD ≥20 vs. PD <20	1.25 (1.01 to 1.54)	1.36 (1.09 to 1.70)
Age at menarche (OR per year)	0.99 (0.92 to 1.06)	1.01 (0.94 to 1.09)
Oral contraceptives (ever/never)	1.09 (0.84 to 1.42)	1.06 (0.82 to 1.39)
Nulliparous (yes/no)	1.28 (0.96 to 1.70)	1.28 (0.97 to 1.71)
Age at first birth (OR per 10 years)	1.02 (0.81 to 1.28)	1.02 (0.81 to 1.28)
Hormone replacement therapy (ever/never)	1.01 (0.81 to 1.26)	1.06 (0.84 to 1.34)
Family history of breast cancer (yes/no)	0.97 (0.74 to 1.28)	0.98 (0.74 to 1.29)

Values in bold represent associations with p values <0.05
BMI body mass index, PD mammographic percent density
^aOutcome: large is > 2 cm, small is ≤ 2 cm
^bAdjusted for age at diagnosis, BMI and PD. All women with complete data on these variables were included

ISSN: 1465-542X