Table 4

Parameter estimates for proportional hazards* and complementary log-log model fitted to the 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) cohort data

ParameterProportional linear hazards ratio modelComplementary log-log discrete time hazards model
EstimateStandard errorEstimateStandard error
Intercept−7.0170.100
log(age/70)2.7790.696
log(age/70)2−3.7821.491
Cohort 2 (1922–<1933)−0.3960.165−0.3860.162
Cohort 3 (1933–<1944)−0.3950.257−0.3960.252
Cohort 4 (1944+)−0.7490.573−0.7170.563
log(TCDD exposure score)†0.0430.0130.0410.013
  • *Proportional hazards regression model of the form, λ  =  λ0 (1+ βχ), adjusted for year of birth cohort (quartiles) and age (the primary time scale for the Cox model).

  • †Cumulative TCDD score under a 15-year lag.