Table 2.
Association between uric acid tertile and high frequency hearing thresholds
| Modela) | Uric acid tertileb) | High frequency hearing threshold |
Low frequency hearing threshold |
||||
|---|---|---|---|---|---|---|---|
| βc) (standard error) | P-value | P for trend | βc) (standard error) | P-value | P for trend | ||
| Unadjusted | T2 vs. T1 | 0.098 (0.012) | <0.001 | <0.001 | 0.031 (0.011) | 0.005 | <0.001 |
| T3 vs. T1 | 0.149 (0.012) | <0.001 | 0.063 (0.011) | <0.001 | |||
| Model 1 | T2 vs. T1 | -0.062 (0.028) | 0.026 | 0.036 | -0.056 (0.028) | 0.046 | 0.080 |
| T3 vs. T1 | -0.067 (0.029) | 0.023 | -0.058 (0.030) | 0.054 | |||
| Model 2 | T2 vs. T1 | -0.068 (0.028) | 0.014 | 0.007 | -0.062 (0.028) | 0.028 | 0.024 |
| T3 vs. T1 | -0.086 (0.030) | 0.005 | -0.074 (0.031) | 0.017 | |||
| Model 3 | T2 vs. T1 | -0.065 (0.028) | 0.020 | 0.008 | -0.065 (0.028) | 0.023 | 0.024 |
| T3 vs. T1 | -0.084 (0.030) | 0.005 | -0.075 (0.031) | 0.015 | |||
T1, ≤4.5; T2, 4.51–5.80; T3, >5.80.
a)
Adjusted covariates: Model 1=age, sex, race; Model 2=Model 1+(triglyceride, high-density lipoprotein cholesterol, body mass index, systolic blood pressure); Model 3=Model 2+(history of smoking)+(heart disease, diabetes mellitus, stroke, noise exposure).
b)
Subjects in the lowest tertile of uric acid were the reference group.
c)
Regression (β) coefficient can be elucidated as change of log-transformed mean high tone hearing thresholds comparing subjects in the upper 2 tertiles of serum uric acid to those in the lowest tertile.