Table 4.
Predicting diabetes based on MetS.
| HR | SE | Wald statistic |
P value | 95% CIs | ||
|---|---|---|---|---|---|---|
| Lower | Upper | |||||
| MetS-based model | ||||||
| Age (years) | 1.00 | 0.01 | 0.93 | 0.354 | 0.99 | 1.01 |
| Family history of diabetes | 1.61 | 0.17 | 4.36 | 0.000 | 1.30 | 1.99 |
| Current smoker | 1.38 | 0.23 | 1.94 | 0.053 | 1.00 | 1.90 |
| Leisure time physical activity | 0.91 | 0.06 | -1.58 | 0.114 | 0.80 | 1.02 |
| BMI (kg.m-2) | 1.05 | 0.01 | 4.07 | 0.000 | 1.03 | 1.07 |
| PCPG (mmol.l-1) | 1.73 | 0.06 | 17.20 | 0.000 | 1.63 | 1.84 |
| MetS | 1.96 | 0.26 | 5.04 | 0.000 | 1.51 | 2.54 |
| Nam-D'Agostino X2 (P for lack of fit) | 8.6 (0.479) | |||||
| Harrell's C (95% CIs) | 0.820 (0.797-0.843) | |||||
| Akaike information criterion | 2812 | |||||
| Bayesian information criterion | 2872 | |||||
BMI, body mass index; HR, hazard ratio; MetS, metabolic syndrome; PCPG, 2-hour post-challenge plasma glucose.