The occurrence of acute episodes of air pollution in the 1950s demonstrated to the general public the harmful effects that deteriorating air quality could have on health. Since then, although pollution levels have dropped significantly—from several hundred to a few dozen micrograms per cubic meter—numerous studies conducted in North America and Europe have shown that exposure to particulate air pollution remains linked to increased mortality and morbidity, particularly of cardio-respiratory origin. Nevertheless, questions have been raised regarding the interpretation of the results, particularly for the effects observed in the short term. The objective of this review is to present the relationship between recent epidemiological findings and their use in calculating the health impact of air pollution. The main elements of causality assessment typically used in epidemiology are discussed in light of current scientific knowledge. An examination of all these elements supports a causal relationship between air pollution and health, and thus justifies the use of relative risks in health impact assessment calculations. Recent studies on the estimation of relative risks show that, even if the excess risk associated with a deterioration in air quality is low—on the order of 1% for a 10 mg/m³ increase—the public health consequences are not negligible due to the size of the exposed population. It is now necessary to make this information accessible to decision-makers and the public so that, with the support of public health professionals, they can contribute to improving urban air quality.