Table 3. Breast cancer incidence risk estimates.

Risk factor	Cases n=109	PYR n=270289	RRa	95% CI
Ethnicity
Slavs	78	165456	1.0	Referent
Tartars & Bashkirs	31	104833	0.7	0.4–1.0
P for homogeneity 0.06, d.f.=1
Subcohort
OTRCb	82	221848	1.0	Referent
Late entrants	27	48441	1.7	1.03–2.6
P for homogeneity 0.04, d.f.=1
Attained age, years:
<45	12	110730	0.1	0.1–0.2
45–54	33	52585	1.0	Referent
55–64	34	50453	1.3	0.8–2.2
65+	30	56521	1.7	1.0–3.0
P for homogeneity <0.001, d.f.=3
Number of children born c
3+	36	124060	1.0	Referent
1–2	53	104546	1.8	1.1–2.7
Nulliparous	19	36483	2.2	1.2–3.8
P for homogeneity 0.008, d.f.=2
Age at first pregnancy c
<20 years	8	27123	1.0	Referent
⩾20 years	81	200998	1.3	0.7–3.0
P for homogeneity 0.43, d.f.=1

^aRelative risk adjusted for age at diagnosis, ethnicity, number of children, age at first pregnancy, time of arrival on the contaminated territory (before and after 1953), linear birth cohort effect and 5-year lagged cumulative stomach dose based on linear ERR model.

^bOTRC=Original Techa River Cohort (resident in the Techa area in 1950–1952).

^cOne case with unknown information on live birth and age at first pregnancy is excluded from analysis.