Causal inference in environmental epidemiology: the role of implicit values

The current epistemologic debate in epidemiology has underlined the relevance of subjective judgment in the production and evaluation of epidemiologic evidence. The definition of criteria aimed at evaluating causal links requires the adoption of judgment, values and extra-scientific considerations, such as the inclusion of a precautionary principle. The purpose of the present analysis is to examine the influence of moral principles in the process of evaluating scientific data relevant to human health. Two case studies are discussed. The first one deals with the carcinogenic risk associated with occupational exposure to non-arsenical insecticides; the second one deals with the association between urban air pollution, mortality and asthmatic attacks.