Table 2.
Association of plasma boron concentrations with all-cause mortality using Cox proportional hazards regression models
| Overall sample | Tertiles | |||
|---|---|---|---|---|
| 1a | 2 | 3 | ||
| Individuals, n (%) | 863 (100.0%) | 287 (33.3%) | 288 (33.4%) | 288 (33.4%) |
| Deceased, n (%) | 99 (11.5%) | 26 (9.1%) | 24 (8.3%) | 49 (17.0%) |
| Plasma boron concentrations in µg/L | 31.9 [22.9; 43.5] | 20.5 [17.1; 22.9] | 31.9 [28.7; 35.2] | 50.9 [43.5; 60.5] |
| HR [95% CI] for all-cause mortality associated with plasma boron | ||||
|---|---|---|---|---|
| 5-unit-increment | ||||
| Model 1 | 1.07 [1.03–1.11] | Ref | 0.86 [0.49–1.50] | 1.85 [1.15–2.97] |
| Model 2 | 1.03 [0.99–1.07] | Ref | 0.57 [0.32–1.00] | 1.01 [0.62–1.66] |
| Model 3 | 1.03 [0.99–1.07] | Ref | 0.63 [0.35–1.12] | 1.14 [0.68–1.93] |
HR are provided for the second and third tertile as compared to the bottom tertile, and per 5-unit-increment in plasma boron. Associations are in hazard ratios [95% confidence interval] with bold figures indicating significant associations (p < 0.05). Model 1: unadjusted. Model 2: adjusted for sex and age. Model 3: Model 2 further adjusted for body mass index, waist-to-hip ratio, education, smoking status, and physical activity
CI confidence interval, HR hazard ratio
aTertile 1 was set as reference