Table 2.
Multiple regression model (linear logistic regression) displaying odds ratios which indicate associations between biomarker principal components and MHO, compared to MUO.
| Model | PP5 | PL2 | ||
|---|---|---|---|---|
| p | OR | p | OR | |
| Model 1 (age + sex) | 0.01∗ | 0.85 | 0.02∗ | 1.06 |
| Model 1 + systolic blood pressure | 0.01∗ | 0.85 | 0.02∗ | 1.06 |
| Model 1 + plasma glucose | 0.02∗ | 0.86 | 0.06 | 1.05 |
| Model 1 + HDL cholesterol | 0.06 | 0.88 | 0.35 | 1.03 |
| Model 1 + triglycerides | 0.04 | 0.87 | 0.24 | 1.04 |
| Model 1 + waist circumference | 0.07 | 0.91 | 0.08 | 1.04 |
| Model 1 + metabolic syndrome | 0.17 | 0.91 | 0.91 | 1.00 |
Odds ratios (OR) indicate associations between biomarker principal components and MHO (=1), compared to MUO (=0). Model 1 was adjusted for age and sex. PL2: lipidomic principal component 2. PP5: proteomic principal component 5. The model adjusted for metabolic syndrome was adjusted for all factors of the metabolic syndrome according to National Cholesterol Education Program panel III (NCEP III) criteria [30] (systolic blood pressure, fasting plasma glucose, HDL cholesterol, triglycerides, and waist circumference). ∗Significant at p < 0.05.