Table 3.
Unconditional logistic regression analysis of MetS/central obesity risks according to the tertiles of serum ZAG levels.
| ZAG tertiles | |||
|---|---|---|---|
| Measurement | Lowest OR (95% CI) | Median OR (95% CI) | Highest OR (95% CI) |
| ZAG in MetS vs. control group | |||
| Range (μg/mL) | <7.60 | ≥7.60–8.60 | ≥8.60 |
| MetS/controls | 58/15 | 51/23 | 42/32 |
| Univariate (R2 = 0.055) | 2.946 (1.419–6.117) | 1.689 (0.861-3.314) | 1 (reference) |
| P-value | 0.004 | 0.127 | |
| Model 1 (R2 = 0.089) | 2.963 (1.413–6.211) | 1.744 (0.879–3.461) | 1 (reference) |
| P-value | 0.004 | 0.112 | |
| Model 2 (R2 = 0.392) | 4.217 (1.716–10.365) | 1.815 (0.804–4.100) | 1 (reference) |
| P-value | 0.002 | 0.151 | |
| Model 3 (R2 = 0.552) | 6.124 (1.992–18.830) | 3.126 (1.076–9.086) | 1 (reference) |
| P-value | 0.002 | 0.036 | |
| ZAG in central obesity vs. control group | |||
| Range (μg/mL) | <7.72 | ≥7.72 to 8.72 | ≥8.72 |
| Central obesity/controls | 33/18 | 25/27 | 26/25 |
| Univariate (R2 = 0.028) | 1.763 (0.796–3.902) | 0.890 (0.411–1.928) | 1 (reference) |
| P-value | 0.162 | 0.768 | |
| Model 1 (R2 = 0.052) | 1.826 (0.817–4.079) | 0.930 (0.425–2.035) | 1 (reference) |
| P-value | 0.142 | 0.855 | |
| Model 2 (R2 = 0.187) | 1.911 (0.806–4.533) | 0.935 (0.402–2.171) | 1 (reference) |
| P-value | 0.142 | 0.875 | |
| Model 3 (R2 = 0.333) | 2.178 (0.794–5.974) | 1.099 (0.406–2.978) | 1 (reference) |
| P-value | 0.130 | 0.852 | |
ZAG, zinc-a2-glycoprotein; MetS, metabolic syndrome. Multivariate odds ratios (OR) and 95% confidence intervals (CI) from unconditional logistic regression models were used in the analysis. Model 1: adjusted for age and gender. Model 2: adjusted for Model 1+ alkaline phosphatase, aspartate transaminase, creatinine, uric acid, urea nitrogen. Model 3: adjusted for Model 2+ grip strength, smoking, drinking, birth place, current address, education level, manual labor, and exercise frequency. Bold font indicated P < 0.05.