Table 4. Models of EOR/Gy and interaction terms for pre-screening and a combined set of pre-screening and prevalent thyroid cancers.

	Pre-screening cases				Pre-screening and prevalent cases
Modifying factor and its values	Cases	EOR/Gya	95% CI	P-valueb	Cases	EOR/Gya	95% CI	P-valueb
All subjects	48	8.18	2.11–72.71		133	3.16	1.49–6.95
Gender
Men	26	2.68	NE–74.59	0.88	62	6.27	1.96–30.44	0.13
Women	22	9.09	1.55–103.30		71	2.06	0.75–5.31
Age at exposure, years
0–4	23	4.98	0.91–43.94	0.32	49	2.82	1.04–7.27	0.92
5–7	12	10.44	2.56–76.99		24	3.33	1.14–8.37
8–18	13	12.24	1.88–109.40		60	3.43	1.23–8.92
Oblast of residence at the time of the accident
Minsk and other	6	10.09	2.70–63.26	0.15	15	1.00	NE–4.19	0.04
Gomel	42	3.41	NE–35.40		118	3.93	1.85–8.65

Abbreviations: CI=confidence interval; EOR/Gy=excess odds ratio per gray; NE=not estimated.

Data restricted to subjects with doses <5 Gy.

^aModel adjusted for age at risk, gender and oblast of residence (Minsk, Gomel, Other).

^bP-value from the likelihood ratio test of homogeneity across categories.