Table 4. Association between railway noise exposure at time of SF-36 and PCS/MCS.
| Model 1 a Estimate (95% CI) | Model 2 b Estimate (95% CI) | Model 3 c Estimate (95% CI) | |
|---|---|---|---|
| Physical Component Summary (PCS) | |||
| Not exposed | 0.00 (ref) | 0.00 (ref) | 0.00 (ref) |
| < 55 dB | -0.06 (-0.30, 0.18) | 0.02 (-0.21, 0.26) | 0.08 (-0.15, 0.31) |
| ≥ 55 dB | -0.42 (-0.75, -0.09) | -0.17 (-0.49, 0.16) | -0.17 (-0.49, 0.14) |
| Linear trend, per 10 dB | -0.26 (-0.49, -0.03) | -0.12 (-0.35, 0.11) | -0.15 (-0.38, 0.07) |
| Mental Component Summary (MCS) | |||
| Not exposed | 0.00 (ref) | 0.00 (ref) | 0.00 (ref) |
| < 55 dB | -0.26 (-0.50, -0.01) | -0.16 (-0.40, 0.09) | -0.12 (-0.36, 0.13) |
| ≥ 55 dB | -0.56 (-0.88, -0.21) | -0.33 (-0.67, 0.00) | -0.32 (-0.66, 0.01) |
| Linear trend, per 10 dB | -0.18 (-0.41, 0.06) | -0.08 (-0.31, 0.15) | -0.10 (-0.33, 0.13) |
aAdjusted for age, sex
bAdjusted for Model 1plus education, cohabitance status, income and railway noise
cAdjusted for Model 2 plus smoking status, waist circumference, and alcohol intake
We investigated effect modification for MSC and PSC by sex, age, education, Charlson Comorbidity Index, and train noise exposure, but found no significant interactions (all p ≥ 0.14, results not shown).