Table 3.

Hazard ratios and 95% Confidence intervals of incident diabetes of TyG index variability (SD).

	Case/Total	Incidence rate, per 1000 person-years	Model 1	Model 2	Model 3
Q1	891/12003	8.80	1.00	1.00	1.00
Q2	933/12003	9.18	1.05(0.96,1.15)	1.04(0.95,1.14)	1.04(0.95,1.14)
Q3	1042/12004	10.28	1.19(1.09,1.30)	1.16(1.06,1.27)	1.13(1.03,1.24)
Q4	1189/12003	11.70	1.39(1.27,1.51)	1.34(1.23,1.47)	1.18(1.08,1.29)
1-SD increase (0.22)			1.20(1.09,1.15)	1.11(1.08,1.15)	1.04(1.01,1.07)
P for Trend			<0.0001	<0.0001	<0.0001

Model 1: adjusted for age (continuous variable, years) and sex (categorical variable, men or women) in 2010.

Model2: included variables in model 1 and further LDL-C (as a continuous variable), HDL-C (as a continuous variable), hs-CRP (as a continuous variable), BMI (as a continuous variable), smoking status (as a categorical variable, yes or no), alcohol consumption status (as a categorical variable, yes or no), physical exercise habits (as a categorical variable, yes or no), educational level (as a categorical variable, high school or above vs. below high school level), hypertension (as a categorical variable, yes or no), and the use of lipid-lowering drugs (as a categorical variable, yes or no) in 2010.

Model 3: included variables in model 2 and further the TyG index (continuous variable) in baseline.