Table 2.

Multivariable regression analysis for the relationship of plasma ALT levels with plasma sE‐selectin or sVCAM levels in the Cohort on Diabetes and Atherosclerosis Maastricht study

Modela	log plasma sE‐selectin (n = 571)			log plasma sVCAM (n = 571)
	β	95% CI	P value	β	95% CI	P value
1	0.346	0.271;0.422	<.001	0.112	0.071;0.152	<.001
2	0.286	0.205;0.367	<.001	0.099	0.055;0.143	<.001
3	0.264	0.177;0.351	<.001	0.101	0.055;0.147	<.001
4	0.248	0.159;0.338	<.001	0.065	0.020;0.110	.005
5	0.249	0.160;0.339	<.001	0.066	0.021;0.111	.004

Analysed with linear regression. Betas represent unstandardized regression coefficients.

Model 2: adjusted for model 1 + BMI, smoking and alcohol intake.

Model 3: adjusted for model 2 + cardiovascular risk factors (total cholesterol, HDL‐cholesterol, systolic blood pressure, HbA1c, use of glucose‐lowering medication, and use of lipid‐modifying medication).

Model 4: adjusted for model 3 + inflammatory markers (C‐reactive protein, amyloid A, interleukin‐6, interleukin‐8, and tumour necrosis factor α).

Model 5: adjusted for model 4 + history of cardiovascular disease.

^aModel 1: adjusted for age + sex.