Table 3.

Multivariate analyses^∗ for prediction of total mortality by oral infection variables and hs-CRP in diabetes and nondiabetes. A 12 1/2-year follow-up of Oslo men of the Oslo II study of 2000.

	Tooth extraction any cause	Tooth extraction any cause, group∗	Tooth extraction infection (TE)∗	Periodontitis (PI)∗	Oral infections (OI)∗	hs-CRP mmol/l	hs-CRP mmol/l, quartiles∗
	HR∗ (95% CI)∗ p value	HR (95% CI) p value	HR (95% CI) p value	HR (95% CI) p value	HR (95% CI) p value	HR (95% CI) p value	HR (95% CI) p value
Diabetes
Oral infection variable or hs-CRP	1.010	1.048	0.923	1.092	0.941	1.022	1.236
	(0.995–1.025)	(0.940–1.169)	(0.684–1.245)	(0.717–1.664)	(0.693–1.277)	(1.011–1.034)	(1.061–1.439)
	0.210	0.399	0.600	0.682	0.695	<0.001	0.006
Nondiabetes
Oral infection variable or hs-CRP	1.007	1.014	1.094	1.016	1.119	1.006	1.086
	(1.001–1.013)	(0.975–1.055)	(0.991–1.2079)	(0.835–1.237)	(1.012–1.237)	(1.004–1.009)	(1.131–1.243)
	0.023	0.488	0.075	0.874	0.029	<0.001	<0.001

^∗All analyses are adjusted for age, years of education, and daily smoking. 95% CI refers to 95% confidence interval; HR: hazard ratio; OI = TE and/or PI, group = stratified groups of numbers of extractions of any cause: 1 = 0-1, 2 = 2–4, 3 = 5–8, 4 = 9–31, 5 = 32, quartiles = analyses by quartile values of hs-CRP.