Air pollution and health: Correlation or causality? The case of the relationship between exposure to particles and cardiopulmonary mortality

Many epidemiologic studies have observed, in different contexts, a slight s hort-term relationship between particles in air and cardiopulmonary mortali ty, even when air quality standards were respected. The causality of this r elationship is important to public health because of the number of people e xposed. Our aim was to make a critical assessment of the arguments used in 15 reviews of published studies. We explain the importance of distinguishing validity from causality, and we systematically analyze the various criteria of judgment within the context of ecologic time studies. Our conclusion is that the observed relationship is valid and that most of the causality criteria are respected. It is hope d that the level of exposure of populations to these particles be reduced. In Europe, acting at the root of the problem, in particular on diesel emiss ions, will also enable the reduction of levels of other pollutants that can have an impact on health. In the United States, the situation is more comp licated, as particles are mainly secondary. It is also essential to continu e with research to become better acquainted with the determinants of person al global exposures and to better understand the toxic role of the various physicochemical factors of the particles.