Table 3—

Pollutant regression models

FeatureKCOFEV1FEV1/FVCUZVILZVI
B-valuep-valueB-valuep-valueB-valuep-valueB-valuep-valueB-valuep-value
Primary particles μg·m−31.060.020.010.050.010.03-1.100.01-1.450.02
NO2 μg·m−30.510.010.0040.060.0030.052-0.410.02-0.450.06
AOT40 μg·m−3·h−1-0.0010.05-5.90×10−60.19-6.25×10−60.030.0010.020.0010.03
SO2 μg·m−30.940.110.0020.740.0060.16-1.140.01-1.230.01
Age-0.48<0.01-0.0030.03-0.003<0.010.41<0.010.68<0.01
Pack-yrs-0.41<0.01-0.01<0.01-0.004<0.010.190.050.48<0.01
Sex-0.730.31-0.030.08-0.06<0.017.30<0.014.570.04
Occupation-0.950.54-0.0020.900.010.38-1.510.28-0.320.84
Carstairs index0.070.87-0.0030.53-0.0010.82-0.360.400.390.41
  • The table shows the regression coefficients and significance of each variable. Models for forced expiratory volume in 1 s (FEV1) and FEV1/forced vital capacity (FVC) used log values while those for transfer coefficient (KCO) and computed tomography densitometry used untransformed data. Results for covariates are from models prior to addition of pollutants. Sex was coded as: male = 1, female = 0. UZVI: upper zone voxel index; LZVI: lower zone voxel index. AOT40 represents accumulated ozone dose, being the sum of the differences between the hourly mean and 40 ppb ozone concentrations for daylight hours exceeding this limit.