Table 4.

Multivariate models with log (DNA adducts) as dependent variable and air pollutants, age, gender, education level, country, and batch as independent variables *
PollutantTime periodSlope P
O31990-19940.0660.0095
PM101992-19940.0280.09
NO21990-19940.0130.17
SO21990-1994−0.0120.048
O31995-19990.00250.26
PM101995-1999−0.0020.35
NO21995-1999−0.0020.32
SO21995-19990.0070.15
Model of logistic regression: DNA adducts (detectable/nondetectable) as dependent variable and air pollutants, age, gender, education level, country, and batch as independent variables
OR § (95% CI)P
O31.97 (1.08-3.58)0.02
PM104.62 (0.43-49.6)0.21
NO20.90 (0.50-1.61)0.71
SO21.12 (0.68-1.83)0.66
  • * DNA adduct measurement preceding air pollution estimates have been deleted.

  • Increase in log adduct level per unit increase of pollutant (μg/m3).

  • Air pollution estimates refer to the same year of blood collection for adduct measurement. Controls only (n = 928).

  • § Adjusted OR comparing the upper tertile versus the lower two tertiles.