Table 4.

Category model^a

	Stomach	Liver	Lung
gen	M/F	M/F	M/F
φmb	11.89	36.90	13.60
Dφ > 0	1311	231	506
T1c	505007.81	104710.40	452429.89
T2	433148.43	100577.71	389939.77
T3	453123.28	97629.42	408114.01
LRTcat-cond	21.59	33.84	19.30
β1e	0.48 (0.21, 0.81)	1.30 (0.68, 2.15)	0.96 (0.17, 2.46)
β2	0.16 (-0.08, 0.46)	0.85 (0.25, 1.75)	1.84 (0.66, 4.08)
β3	0.42 (0.16, 0.75)	1.39 (0.70, 2.34)	1.33 (0.38, 3.11)
Dev	1661.13	1235.38	1035.39
Chi-sq	3391.69	2451.56	3111.30
df	2990	2990	2990
	Colon	Uterus	All solid
gen	M/F	F	M/F
φmb	20.28	26.91	43.86
Dφ > 0	184	127	365
T1c	342887.98	128431.04	24656.43
T2	288767.71	153667.68	21860.18
T3	307331.28	145793.72	21917.87
LRTcat-cond	1.57	8.08	16.31
β1e	0.07 (-0.40, 0.95)	0.92 (0.19, 2.08)	0.13 (-0.06, 0.34)
β2	0.32 (-0.31, 1.52)	0.28 (-0.30, 1.29)	0.43 (0.16, 0.74)
β3	0.02 (-0.44, 0.89)	0.52 (-0.08, 1.48)	0.29 (0.06, 0.56)
Dev	718.42	630.89	2121.74
Chi-sq	3625.26	1038.26	2772.40
df	2990	1511	2990

^a 0 – 20 mSv subcohort, Excess Relative Risk ERR = β₁dosecat₁ + β₂dosecat₂ + β₃dosecat₃ with β_i constrained by 1 + β_i ≥ 0; dosecat₁ = 1 if 0 < D_φ ≤ 0.025, and = 0 otherwise; dosecat₂ = 1 if 0.025 < D_φ ≤ 0.05, and = 0 otherwise; dosecat₃ = 1 if 0.05 < D_φ, and = 0 otherwise. The Control model is nested by β₁ = β₂ = β₃ = 0.

^b Latency φ_m is chosen from the results of the two-phase model (Table 3)

^c At φ_m, T_j is the p-y observation for which dosecat_j = 1, j = 1,2,3

^d At φ_m, the likelihood ratio test LRT_cat-con compares the category model with the control model (χ² on 3 df)

^e At φ_m, β_i are the fitted Maximum Likelihood Estimate parameter values, with 95%CI's

Other definitions as in Table 3.