The German Ausschuss für Gefahrstoffe (AGS: Committee for dangerous substances) has laid down a socio-political concept for the derivation of health-based occupational limit values for carcinogens (so-called AGSLeitfaden). The described procedure and the examples of its application suffer from an essential methodological error: the calculated excess risks of dying are meaningless because every human being must die once and there can be no “excess deaths due to exposure”. On the other hand, if the AGS-Leitfaden were to deal strictly with diseases and not with causes of death, the constant excess risk levels (“tolerable”, “acceptable”) set by the AGS would have to be varied with the severity and frequency of the diseases. In any case the AGS-Leitfaden should be revised. I recommend as an alternative method of evaluating exposure to dangerous substances the determination of “years of life lost“. This would not only solve the methodological problems mentioned above but would also open up a new perspective for evaluation. The personal life expectancy offers a realistic benchmark for the evaluation of potential effects of exposure reductions (“years of life saved”) whereas an abstract probability difference cannot be interpreted in a similar way.
