摘要：Author Georges Salines InVS Département santé environnement 12, rue du Val d'Osne 94415 Saint-Maurice cedex France Key words: bias (epidemiology), environmental exposure, epidemiology, low doses, risk assessment, uncertainty DOI : 10.1684/ers.2010.0379 Page(s) : 431-4 Published in: 2010 This paper is a commentary of André Aurengo's article entitled “Epidemiology: the Age of Doubt”, published in issue #4 (July-August 2010) of Environnement, Risques et Santé. The author affirmed in that article that epidemiology has flaws that tend to bias its results toward false-positive outcomes. He expressed concerns about consequences of this phenomenon on decision making and suggested as a possible solution the development of a scale of validity for epidemiological studies. This point of view appears to lack sufficient supporting arguments and examples. We offer a more balanced evaluation of epidemiology that points out the existence of methodological shortcomings that prevent conclusions that a risk exists, even when it actually does. The fear expressed by Aurengo that faulty decision making in public health matters will follow from isolated and biased studies seems exaggerated in relation to the current context and at odds with the real issues of risk perception and decision making in situations of uncertainty. Finally, the relevance of the proposed solution is discussed.
