Buccal mucosa micronuclei counts in relation to exposure to low dose-rate radiation from the Chornobyl nuclear accident and other medical and occupational radiation exposures

Table 3 Regression analysis of radiation-associated absolute and relative risk of micronucleus prevalence (after exclusion of 6 cancer cases)

Model number	Linear term for absolute risk (α) model (/1000 Scored cells /Gy)	Linear term for relative risk (α) (/Gy)	Exponential term (β) (/Gy)	Deviance (df)	p-value
	AIC-minimizing background model (no dose term)
1	-	-	-	106.04 (75)
	Absolute risk model in dose (1), background adjusted using AIC-minimizing model
2	3.03 (−0.78^a, 7.65)	-		102.56 (74)	0.1170^b
3	0.71 (−3.05^a, 16.40)	-	2.51 (−5.36, 11.05^a)	102.10 (73)	0.5710^c
	Relative risk model in dose (2), background adjusted using AIC-minimizing model
4	-	1.19 (−0.49, 3.93)	-	103.60 (74)	0.1902^b
5	-	0.14 (−0.97^a, 7.98)	4.27 (−8.36^a, 16.89^a)	102.67 (73)	0.4205^d

All p-values relate improvement in fit evaluated via an F-test. All models adjust for background variables that minimize Akaike Information Criterion (AIC), as in Appendix 1 Table 7. 2-sided p-values and confidence intervals are adjusted for overdispersion
^aWald-based CI
^b p-value of improvement in fit over null model in dose (model number 1)
^c p-value of improvement in fit over linear model in dose (model number 2)
^d p-value of improvement in fit over linear model in dose (model number 4)

ISSN: 1476-069X