Table 3.
Multiple linear regression results using indicators of MS as continuous variables
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| Predictor | β | p-value | β | p-value | β | p-value |
| WC | 0.069 | 0.267 | 0.026 | 0.702 | −0.111 | 0.236 |
| BPsys† | −0.087 | 0.288 | 0.041 | 0.644 | 0.036 | 0.679 |
| BPdia | 0.113 | 0.167 | 0.020 | 0.815 | 0.007 | 0.935 |
| HDL† | 0.094 | 0.154 | 0.107 | 0.103 | 0.116 | 0.076 |
| TG† | 0.016 | 0.811 | 0.067 | 0.318 | 0.073 | 0.273 |
| FBG* | −0.072 | 0.228 | −0.043 | 0.480 | −0.051 | 0.397 |
†Indicates that log-transformed values are used in the analysis. *indicates that dichotomous values (0=absence/1=presence of elevated FBG) are used in the analysis.
Model 1 includes indicators of metabolic phenotype. Model 2 includes predictors in model 1, age, ethnicity, smoking status, total testosterone and sex hormone-binding globulin levels. Model 3 includes predictors in model 2 and body mass index.
Abbreviation: BMI body mass index, BP blood pressure, FBG fasting blood glucose, HDL high density lipoprotein cholesterol, MS metabolic syndrome, TG triglycerides, WC waist circumference, β standardized coefficient of regression.