Table 4 Odds ratios from logistic regression analysis for exposures assigned by occupational hygienists to the occupation in the 3 months prior to semen sample
ExposurenUnadjustedAdjusted*
CasesReferentsOR (95% CI)OR (95% CI)
Glycol ethers
    No65394911
    Low1522420.91 (0.73 to 1.14)0.93 (0.74 to 1.17)
    Moderate48411.70 (1.11 to 2.61)1.65 (1.06 to 2.54)
    High21122.54 (1.24 to 5.21)2.54 (1.23 to 5.27)
All other VOCs
    No55982111
    Low2113250.95 (0.78 to 1.17)0.97 (0.79 to 1.19)
    Moderate76731.53 (1.09 to 2.15)1.55 (1.10 to 2.18)
    High28251.65 (0.95 to 2.85)1.68 (0.96 to 2.93)
Carbon disulphide
    No824118211
    Low46601.10 (0.74 to 1.63)1.13 (0.76 to 1.69)
    Moderate322.15 (0.36 to 12.91)3.31† (0.59 to 18.43)
    High10
Styrene
    No786113311
    Low71961.07 (0.77 to 1.47)1.10 (0.79 to 1.52)
    Moderate12131.33 (0.60 to 2.93)1.43 (0.64 to 3.18)
    High523.60 (0.70 to 18.62)3.12 (0.58 to 16.64)
  • OR, odds ratio; VOC, volatile organic compound.

  • n = 2118.

  • *In a multi-level logistic regression model including: surgery to the testes, previous conception, use of alcohol and wearing boxer shorts, clustered within centre.

  • †“High” combined with “moderate” to eliminate empty cell.