Table 3.
Multivariate Cox-proportional hazards model (Breslow method for ties) showing a significantly increased mortality rate among individuals allocated to the fourth quartile of neck circumference compared with those in lower quartiles. In this model, increased age at baseline, poor physical activity and high fasting glucose were directly associated with mortality, while a high body mass index was inversely associated
| Neck circumference | HR | 95% CI | p value |
|---|---|---|---|
| Quartile 1 (25–34 cm) | Referent category | ||
| Quartile 2 (35–36 cm) | 1.24 | 0.83–1.86 | 0.297 |
| Quartile 3 (37–39 cm) | 1.39 | 0.89–2.16 | 0.891 |
| Quartile 4 (40–50 cm) | 2.98 | 1.77–5.02 | <0.001* |
| Age at baseline | 1.06 | 1.04–1.07 | <0.001* |
| Being female | 1.20 | 0.83–1.74 | 0.325 |
| Primary school education | 1.17 | 0.81–1.70 | 0.409 |
| Current smoker | 0.69 | 0.28–1.70 | 0.420 |
| Body mass index ≥30 kg/m2 | 0.58 | 0.39–0.84 | 0.005* |
| Poor physical activity | 2.34 | 1.63–3.35 | <0.001* |
| Poor diet | 0.96 | 0.56–1.65 | 0.895 |
| Blood pressure ≥140/90 mmHg | 1.13 | 0.85–1.50 | 0.395 |
| Fasting glucose ≥126 mg/dL | 1.92 | 1.45–2.54 | <0.001* |
| Total cholesterol ≥240 mg/dL | 0.59 | 0.35–1.01 | 0.055 |
*Statistically significant result.